ArticlePDF Available

A Phenotypic Comparison of Loudness and Pain Hyperacusis: Symptoms, Comorbidity, and Associated Features in a Multinational Patient Registry


Abstract and Figures

Purpose: Hyperacusis is a complex and poorly understood auditory disorder characterized by decreased tolerance to sound at levels that would not trouble most individuals. Recently, it has been suggested that individuals who experience otalgia in response to everyday sounds (termed “pain hyperacusis”) may differ clinically from those whose primary symptom is the perception of everyday sounds as excessively loud (termed “loudness hyperacusis”). Despite this theoretical distinction, there have been no empirical studies directly comparing these two populations of hyperacusis patients. Method: Using data from a multinational patient registry (the Coordination of Rare Diseases at Sanford [CoRDS] Registry), we examined self-reported demographics, symptoms, comorbidity, and response to treatment in a sample of 243 adults with hyperacusis, 152 of whom were classified as having pain hyperacusis based on reported symptoms. Bayesian statistical tests were used to investigate both the presence and absence of group differences between patients with loudness and pain hyperacusis. Results: Individuals with pain hyperacusis presented with a more severe clinical phenotype, reporting a higher frequency of temporary symptom exacerbations (i.e., “setbacks”), less perceived symptom improvement over time, more severe comorbid headache disorders, and reduced benefit from sound therapy. However, the two hypothesized hyperacusis subtypes exhibited more similarities than differences, with the majority of symptoms and comorbidities being equally prevalent across groups. Multiple comorbidities were commonly observed, including tinnitus, primary headache disorders, psychiatric disorders, and functional somatic syndromes. Intolerance of sensory stimuli in other modalities was also frequently reported. Conclusion: Although this study provides little evidence that loudness and pain hyperacusis are pathophysiologically distinct conditions, our findings indicate that a pain-predominant phenotype may be a meaningful prognostic marker in patients with hyperacusis.
Content may be subject to copyright.
Research Article
A Phenotypic Comparison of Loudness
and Pain Hyperacusis: Symptoms,
Comorbidity, and Associated Features
in a Multinational Patient Registry
Zachary J. Williams,
Evan Suzman,
and Tiffany G. Woynaroski
Purpose:AQ1 Hyperacusis is a complex and poorly understood
auditory disorder characterized by decreased tolerance to
sound at levels that would not trouble most individuals.
Recently, it has been suggested that individuals who
experience otalgia in response to everyday sounds (termed
pain hyperacusis)maydifferclinicallyfromthosewhoseprimary
symptom is the perception of everyday sounds as excessively
loud (termed loudness hyperacusis). Despite this theoretical
distinction, there have been no empirical studies directly
comparing these two populations of hyperacusis patients.
Method: Using data from a multinational patient registry (the
Coordination of Rare Diseases at Sanford Registry), we
examined self-reported demographics, symptoms, comorbidity,
and response to treatment in a sample of 243 adults with
hyperacusis, 152 of whom were classified as having pain
hyperacusis based on reported symptoms. Bayesian
statistical tests were used to investigate both the presence
and absence of group differences between patients with
loudness and pain hyperacusis.
Results: Individuals with pain hyperacusis presented
with a more severe clinical phenotype, reporting a higher
frequency of temporary symptom exacerbations (i.e.,
setbacks), less perceived symptom improvement over
time, more severe comorbid headache disorders, and
reduced benefit from sound therapy. However, the two
hypothesized hyperacusis subtypes exhibited more
similarities than differences,withthemajorityofsymptoms
and comorbidities being equally prevalent across
groups. Multiple comorbidities were commonly observed,
including tinnitus, primary headache disorders, psychiatric
disorders, and functional somatic syndromes. Intolerance
of sensory stimuli in other modalities was also frequently
Conclusion: Although this study provides little evidence
that loudness and pain hyperacusis are pathophysiologically
distinct conditions, our findings indicate that a pain-
predominant phenotype may be a meaningful prognostic
marker in patients with hyperacusis.
Hyperacusis is a hearing disorder characterized by
decreased tolerance to sound at levels that would
not trouble most people (Fackrell et al., 2019).
Individuals experiencing hyperacusis report that everyday
sounds are perceived as excessively loud, intense, fright-
ening, overwhelming, or physically painful, and can trig-
ger additional physical symptoms, leading to significant
distress and functional impairment (Ke et al., 2020; Tyler
et al., 2014). This condition is distinct from misophonia
(an aversive reaction to specific triggersounds charac-
terized by anger, extreme annoyance, and disgust) and
phonophobia (a persistent, abnormal, and unwarranted
fear of certain sounds), although these other forms of de-
creased sound tolerance can co-occur with hyperacusis
Medical Scientist Training Program, Vanderbilt University School of
Medicine, Nashville, TN
Department of Hearing and Speech Sciences, Vanderbilt University
Medical Center, Nashville, TN
Vanderbilt Brain Institute, Vanderbilt University, Nashville, TN
Frist Center for Autism and Innovation, Vanderbilt University,
Nashville, TN
Graduate Program in Biomedical Sciences, Vanderbilt University,
Nashville, TN
Vanderbilt Kennedy Center, Vanderbilt University Medical Center,
Nashville, TN
Correspondence to Zachary J. Williams:
Editor-in-Chief: Ryan W. McCreery
Editor: Jamie Bogle
Received December 4, 2020
Revision received January 23, 2021
Accepted January 26, 2021
Disclosure: Zachary J. Williams has received consulting fees from Roche. He
also owns stock and stock options in Axsome Therapeutics, Inc., CRISPR
Therapeutics AG, and Editas Medicine. The other authors have declared that no
other competing interests existed at the time of publication. No funding body or
source of support had a role in the study design, data collection, analysis, or
interpretation, decision to publish, or preparation of this article.
American Journal of Audiology 118 Copyright © 2021 American Speech-Language-Hearing Association 1
AJA-20-00209Williams (Author Proof )
(Jastreboff & Jastreboff, 2015). Although prevalence esti-
mates of hyperacusis are varied and highly method depen-
dent, an estimated 5.9% of U.S. adults report problems
tolerating everyday sounds based on data from the 2014
National Health Interview Survey (Zelaya et al., 2015), and
self-reported sound intolerance was endorsed by 9% of adults
in a large Swedish study (Paulin et al., 2016). Hyperacusis
is also associated with a range of other disorders, including
tinnitus, hearing impairment, traumatic brain injury, mi-
graine, anxiety disorders, mood disorders, autism spectrum
disorder, and functional somatic syndromes such as fibro-
myalgia and irritable bowel syndrome (Assi et al., 2018;
Cederroth et al., 2020; Paulin et al., 2016; Sheldrake et al.,
2015; Williams, He, et al., 2021; Williams, Suzman, &
Woynaroski, 2021). At this time, research on hyperacusis
is still in its infancy; the etiology, pathology, and natural
history of this condition remain poorly understood (Baguley
& Hoare, 2018; Pienkowski et al., 2014; Tyler et al., 2014),
and no evidence-based recommendations currently exist to
guide its diagnosis or treatment.
One recent approach to the study of hyperacusis has
been dividing the condition into subtypes based on the spe-
cific aversive reactions to sounds that patients experience.
In their seminal review, Tyler et al. (2014) proposed that
hyperacusis could be separated into four subtypes: loud-
ness hyperacusis, pain hyperacusis, annoyance hyperacusis,
and fear hyperacusis, occurring either separately or in com-
bination. Though many researchers consider misophonia
(i.e., Tylersannoyance hyperacusis) and phonophobia
(i.e., Tylersfear hyperacusis) to be separate conditions
from hyperacusis (Fackrell et al., 2019; Fagelson & Baguley,
2018; Jastreboff & Jastreboff, 2015), the more narrowly de-
fined hyperacusis can still be subdivided into loudness and
pain subtypes. For individuals with loudness hyperacusis,
the threshold for loudness discomfort is reduced, and sounds
of moderate intensity are judged to be very loud (Phillips
& Carr, 1998). These symptoms are associated with distur-
bances of loudness perception, as evidenced by excessive
growth of perceived loudness as a function of sound inten-
sity (Brandy & Lynn, 1995; Hébert et al., 2013; Noreña &
Chéry-Croze, 2007). Pain hyperacusis is a less well-understood
form of the condition in which an individual perceives phys-
ical pain in the ear when exposed to certain sounds at levels
far below those needed to cause pain in a typical listener
(i.e., approximately 120 dB SPL). The character of the pain
is variable across individuals (e.g., dull ache, burning, sharp,
stabbing, or throbbing), and patients often report experienc-
ing transient symptom exacerbations in response to certain
sounds, colloquially referred to as setbacks(Pollard, 2019).
Although the pathophysiology of pain hyperacusis is still
poorly understood, recent discoveries have implicated the
population of Type II cochlear afferent neurons in the per-
ception of noxious or painful auditory stimuli (Flores et al.,
2015; Liu et al., 2015; Wu et al., 2018). Although specula-
tive, researchers have hypothesized that increased sensitivity
or inappropriate activation of this population of Type II
afferents may mediate the symptoms of pain hyperacusis
(Auerbach, 2019; Noreña et al., 2018). To date, there has
been very little empirical research characterizing the clinical
phenotype of pain hyperacusis, and it remains unclear
whether pain hyperacusis can be meaningfully distinguished
from loudness hyperacusis in terms of clinical severity, nat-
ural history, associated symptoms, comorbidity, or response
to treatment.
Patient self-report is increasingly recognized as an im-
portant modality for assessing the impact of disease states
and generating novel questions for investigation (Acquadro
et al., 2003; Weldring & Smith, 2013). These data are partic-
ularly informative when investigating disorders such as
hyperacusis that are predominantly characterized by sub-
jective symptoms, and large-scale patient surveys have pro-
vided useful insights into several similar conditions (Bennett
et al., 2007; Chu et al., 2018; Kanazawa et al., 2016; Rouw
& Erfanian, 2018). Leveraging an existing database of pa-
tient survey responses (the Coordination of Rare Diseases
at Sanford [CoRDS] Registry; Trudeau, 2013), the current
study was designed to provide an in-depth clinical descrip-
tion of the hyperacusis phenotype, with the specific goal of
examining meaningful differences between loudness and
pain subtypes of the condition. The major aims of this study
were to compare individuals with loudness and pain hypera-
cusis with respect to (a) demographics, background, and on-
set of hyperacusis; (b) quality of life (QoL) and functional
impairment; (c) specific symptoms attributed to hyperacusis;
(d) clinical course and symptom fluctuation (including the
presence and severity of setbacks); (e) comorbid condi-
tions; (f) comorbid nonauditory sensory symptoms; and
(g) perceived benefit of interventions such as medication
and sound therapy.
Method AQ2
Study Sample and Data Source
Data for the current study were drawn from a multi-
national hyperacusis patient registry (hereafter the CoRDS
Registry) established in 2015 by the nonprofit organization
Hyperacusis Research Limited, Inc. in partnership with San-
ford Research (Pollard, 2019; Trudeau, 2013). Anonymized
data from this repository are available from Sanford Re-
search upon request (see Availability of Data and Materials
section for additional details). For the current study, data
were obtained from the CoRDS Registry in October of
2020. The study protocol for the secondary analysis of
these data was approved by the institutional review board
of Vanderbilt University Medical Center.
At the time of data access, a total of 456 individuals
had provided data for this registry, including both adult
patients with self-reported hyperacusis and parents or
caregivers of minors with hyperacusis. For the current
study, we limited our scope to self-reporting adults, given
the subjective nature of hyperacusis symptoms and the
relatively small number of minors in the sample (n= 18
[3.9%]). Included patients were required to (a) self-report
a diagnosis (including self-diagnosis) of hyperacusis in re-
sponse to the question Please list all rare disease diagnoses,
2American Journal of Audiology 118
AJA-20-00209Williams (Author Proof )
(b) report symptoms consistent with either loudness or pain
hyperacusis (i.e., everyday sounds are perceived as overly
loud and/or cause physical pain), and (c) provide an answer
to the question, In the past twelve (12) months, approxi-
mately how often has the Participant experienced pain in one
or both ears?(response options: Never,”“Once a month,
23 times per month,”“Once a week,”“Every day,and
Continuously), which we used to characterize individuals
into the categories of loudness and pain hyperacusis along
with free-text reports of hyperacusis symptomatology. A
total of 249 adults answered the question regarding pain
frequency, and of these, an additional six prospective par-
ticipants were eliminated from the study due to either not
reporting a diagnosis of hyperacusis or reporting no symp-
toms consistent with loudness or pain hyperacusis (i.e., some
patients recorded only strong emotional reactions to sound
[i.e., misophonia] or reactive tinnitus without endorsing
either pain in the ears or the perception of everyday sounds
as excessively loud). Thus, the final sample of participants
analyzed in the current study comprised 243 unique individ-
uals with hyperacusis.
Survey Questions
Participants completed two questionnaires, including
a 79-item background survey completed by patients in all
CoRDS registries and a 90-item hyperacusis-specific survey
designed for this particular registry (full questionnaires avail-
able from CoRDS on request). Analyzed questions from the
background survey included (a) demographic variables (age,
sex, race, ethnicity, country of residence), (b) rare and non
rare disease diagnoses (free text), (c) symptoms attributed
to hyperacusis (free text), (d) and five items assessing vari-
ous facets of QoL (1. In general, would the participant say
his/her health is?; 2. Does the participants health now
limit him/her in doing vigorous activities? ; 3. How much
did pain interfere with the participants enjoyment of life?;
4. How often does the participant feel tired?;5.The partic-
ipant feels depressed.). Participants responded to these
QoL items using a 5-point Likert scale, with the first ques-
tion rated from Excellentto Poorand the other four
rated from Never to Always. Item scores were coded such
that higher scores indicated higher QoL (i.e., a response
of Alwayswas scored as 1and a response of Never
was scored as 5).
The hyperacusis-specific questionnaire was com-
posed of four sections: (a) background and onset of hyper-
acusis (e.g., duration of hyperacusis, laterality, symptom
change over time, hyperacusis risk factors), (b) symp-
toms (e.g., perception of everyday sounds as unbearably
loud, pain character and frequency, associated symp-
toms such as pressure and ear fullness, setback frequency
and duration, and degree of functional impairment from
hyperacusis), (c) comorbid conditions (e.g., presence and
characteristics of tinnitus, headache frequency and severity,
additional medical conditions, presence and severity of
sensory intolerance in other modalities), and (d) hyper-
acusis management and treatment (e.g., response to sound
therapy, use of ear protection, perceived benefit from
medications). Specific outcomes analyzed in the current
study and the survey questions used to assess them are de-
tailed in T1Table 1.
Classification of Loudness and Pain Hyperacusis
As part of the hyperacusis-specific survey, participants
answered the following question: In the past twelve (12)
months, approximately how often has the Participant experi-
enced pain in one or both ears? Notably, only 9.5% of our
sample (n= 23) responded Neverto this question, indi-
cating that the majority of hyperacusis patients do endorse
noise-induced pain at least some of the time. However,
the frequency of pain episodes in the sample varied widely,
with 24.7% of participants (n= 60) endorsing pain once
per week or less, 45.7% (n= 111) endorsing pain daily,
and 20.2% (n= 49) endorsing continuous pain in their ears.
Thus, for an individual to be classified as having pain
hyperacusis, we required that they report ear pain as oc-
curring Every dayor Continuously(n= 160 [65.8%]).
Individuals reporting pain weekly or less were classified
as having loudness hyperacusis. Participants who reported
pain were also surveyed regarding the character of their
pain (with response options of Dull ache,”“Burning pain,
Throbbing pain,”“Sharp pain,”“Stabbing pain,and
Other), and free-text explanations of those who marked
the Otherresponse were examined to determine whether
the sensation being described was clearly indicative of
pain. In cases where the free text unambiguously clari-
fied that the sensation was not pain (e.g., the responses a
nervous panicky reactionand A buzzy distortion), par-
ticipants were removed from the pain hyperacusis group
and classified as having solely loudness hyperacusis. This
process resulted in eight individuals being reclassified as
having loudness hyperacusis rather than pain hyperacusis.
Thus, using this method, 152 individuals in our sample
(62.6%) were classified as having pain hyperacusis, while
91 (37.4%) were classified as having loudness hyperacu-
sis. Notably, those classified as having pain hyperacusis
frequently reported excessive loudness perception, with
over 90% of participants in the pain hyperacusis group
reporting that they find everyday sounds to be unbear-
ably loud,as evidenced by affirmative responses to the
question, Are everyday sounds unbearably loud to the
Coding of Free-Text Responses
In addition to the multiple-choice questions in the
background and hyperacusis-specific surveys, participants
answered several open-ended questions to qualitatively
describe their symptoms (e.g., Please list symptoms of
rare disease diagnosisand List any other primary medical
conditions (disease/condition, autoimmune disorder, cancer,
etc.) that affect the participant). Open-ended questions were
also included to elaborate on responses of Otherprovided
by participants to questions like What other types of
Williams et al.: Comparison of Loudness and Pain Hyperacusis 3
AJA-20-00209Williams (Author Proof )
Table 1. Outcomes of interest derived from closed-ended questions on the hyperacusis-specific CoRDS survey.
Outcome Data type Source question Response options n
Duration of
Ordinal How long has the participant had
Less than 1 year; 13 years;
46 years; 710 years;
1114 years; 15+ years
144 87
Sudden onset Binary (yes/no) How did the onset of the condition
Suddenly = yes
Gradually; uncertain = no
143 85
Bilateral symptoms Binary (yes/no) Which ear is affected by hyperacusis? Both = yes;
Left; right = no
149 90
Symmetric symptoms Binary (yes/no) Are both ears equally affected? Yes; no 150 89
Professional diagnosis Binary (yes/no) How was the participant initially
Audiologist; ENT; medical
doctor = yes
Self; other = no
141 86
Pre-existing hearing loss Binary (yes/no) Does the participant have pre-existing
hearing loss?
Yes; no 152 88
Wears hearing aids Binary (yes/no) Does the participant wear hearing aids? Both ears; left; right; = yes
140 83
Head injury Binary (yes/no) Has the participant ever had a serious
head injury (from car accident,
fall, etc.)?
Yes; no 147 88
Loud noise exposure Binary (yes/no) Does the participant have a history
of loud noise exposures?
No; uncertain = no
150 89
Traumatic noise/blast
Binary (yes/no) Has the participant had traumatic
impulse noise exposures (blasts,
gun fire, etc.)?
Yes; no 151 90
Delayed pain onset Binary (yes/no) When the participant experienced
ear pain as a result of an event,
how long after the event did the
pain begin?
After a few hours; the next day;
a few days later; weeks
later = yes
Immediately = No
131 57
Duration of pain Ordinal When the participant experiences ear
pain from environmental sounds,
how long does the pain last?
Less than 1 hr; 14 hr; 524 hr;
several days; several weeks;
several months; more than
several months
128 56
Functional impairment:
Ordinal Rate the impact of the participants
hyperacusis on the participants
010 (10 indicates the participants
condition prevents them from
going to school or working)
142 88
Functional impairment:
Ordinal Rate the impact of the participants
hyperacusis on the participants
home life.
010 (10 indicates the participants
condition ruined their home life)
147 89
Functional impairment:
Ordinal Rate the impact of the participants
hyperacusis on the participants
social life.
010 (10 indicates the participants
condition ruined their social life)
145 89
Symptom change
over time
Categorical Has the severity of the participants
hyperacusis changed over time?
Better; worse; fluctuated; same 147 87
Setback severity Ordinal When the participant experiences
a setback, how does the participants
condition compare to earlier periods?
Small setback; moderately worse;
worse than it ever was previously
121 69
Setback recovery time Ordinal How long does it take for the participant
to recover from a setback?
Several days; several weeks;
months; years
119 70
Recovery worsened
by sounds
Binary (yes/no) Does the participants recovery take
longer if they are exposed to a
louder noise?
Yes; no 120 73
Tinnitus impairs sleep Binary (yes/no) Does the participant have difficulty
sleeping due to their tinnitus?
Yes; no 126 78
Tinnitus impairs
Binary (yes/no) Does the participant have difficulty
concentrating while reading in a
quiet room due to their tinnitus?
Yes; no 125 78
Reactive tinnitus Binary (yes/no) Do exposure sounds raise the volume
of the participants tinnitus?
Yes; no 122 77
Reactive tinnitus
Ordinal If the volume level of the participants
tinnitus increases, how long does
the increase in volume last?
Minutes; hours; days; weeks;
117 68
Headache frequency Ordinal How often does the participant
experience headaches?
Never; rarely; monthly; weekly;
143 88
Headache severity Ordinal How severe are the participants
headaches? (10 being the most
110 124 70
(table continues)
4American Journal of Audiology 118
AJA-20-00209Williams (Author Proof )
physical sensations does the participant experience along
with hyperacusis?and When the participant experiences
ear pain from environmental sounds, what type of pain do
they experience?
In order to more fully characterize the clinical phe-
notypes of participants in this registry, we examined each
participants free-text responses, classifying them according
to the additional sound-evoked symptoms they reported
(e.g., ear fullness, jaw pain, hearing changes). These re-
sponses were categorized into binary variables that indi-
cated whether a specific symptom was mentioned by a given
patient. A similar procedure was undertaken for the ques-
tions regarding additional medical comorbidities and specific
medications that patients found helpful for their hypera-
cusis symptoms. As the presence of certain comorbidities
(e.g., temporomandibular joint disorders [TMJDs] and
tinnitus) could be indicated in a multiple-choice question
or a free-text field, we coded a given condition as present
if it was endorsed by a patient in either location. Binary
variables were created for any symptom category, comor-
bid condition, or medication class that was reported by
5% or more of the sample (n12; see below paragraph
for a full list). Notably, as only 84 patients provided a re-
sponse to the question regarding medication benefit, we
created categories for medication classes perceived as ben-
eficial by eight or more individuals (i.e., approximately
10% of the individuals who responded to that question).
In the domain of symptoms, we categorized pain char-
acter into three nonexclusive categories based on the latent
structure of the Short-form McGill Pain Questionnaire
(Dworkin et al., 2009): (a) continuous pain (throbbing,
aching,”“gnawing,”“heavy,or tender); (b) intermit-
tent pain (stabbing,”“sharp,”“shooting,”“splitting,
electric-shock,and piercing); and (c) neuropathic-like
pain (hot/burning,”“cold/freezing,”“pain caused by light
touch; note that, while paresthesias are also included in
this section of the Short-form McGill Pain Questionnaire,
we classified such sensations as associated symptoms rather
than pain per se). Participants who described multiple types
of pain in their descriptions were classified in multiple cate-
gories. For participants who described pain but did not re-
spond to the multiple-choice item describing pain character
or provide any descriptors, we considered responses to
the pain character categories to be missing. With regard to
Table 1. (Continued).
Outcome Data type Source question Response options n
Headache worsened
by sounds
Binary (yes/no) Does the participant experience
headaches more frequently after
they have been exposed to loud
Yes; no 122 71.
Vertigo daily/weekly Binary (yes/no) How often does the participant
experience balance problems
Never; rarely; monthly = no
Weekly; daily = yes
144 86.
Vertigo severity Ordinal Rate the severity of the participants
balance issues (10 being the most
110 90 58.
Vertigo worsened
by sound
Binary (yes/no) Does the participant experience
increased/more frequent balance
issues after they are exposed to
loud noises?
Yes; no 91 57.
Photophobia daily/
Ordinal How often do bright lights bother
the participant?
Never; rarely; monthly = no
Weekly; daily = yes 141 83
Photophobia severity Ordinal How severely do bright lights bother
the participant? (10 being the
most severe)
104 61
Osmophobia present Binary (yes/no) Is the participant bothered by strong
Yes; no
140 81
Tactile intolerance
Binary (yes/no) Is the participant bothered by touch? Yes; no
143 85
Duration of sound
Ordinal How long has the participant
followed/utilized sound therapy
Less than 1 year; 12 years;
34 years; 56 years;
78 years; 8+ years
88 48
Outcome of sound
Ordinal How much improvement has the
participant experienced from
following/utilizing sound therapy
Worsened hyperacusis;
worsened tinnitus =
worsened symptoms
No improvement; minor improvement;
some improvement; significant
improvement; hyperacusis
is almost eliminated
91 49
Note. Additional outcomes examined were (a) derived from CoRDS background survey, (b) derived from free-text fields, or (c) derived from a
combination of free-text and multiple-choice fields. Number of nonmissing responses determined both by voluntary skipping (e.g., providing
no answer) and survey logic (e.g., severity of headache not rated by individuals not endorsing headaches); n
= number of patients
in pain/loudness hyperacusis group providing nonmissing response to question. CoRDS = Coordination of Rare Diseases at Sanford.
Williams et al.: Comparison of Loudness and Pain Hyperacusis 5
AJA-20-00209Williams (Author Proof )
ancillary symptoms, the following were reported frequently
enough to warrant categorization: pain in face, jaw, and
neck regions; ear fullness; ear pressure; scalp or head pres-
sure; flu-like symptoms (e.g., fatigue, malaise); fluttering/
middle ear myoclonus; hearing changes (e.g., distortion,
saturation, autophony, temporary hearing loss); sounds
such as clicking or popping; and paresthesia (defined broadly,
including numbness, tingling, heat, cold, and vibrotactile
sensations). Comorbidities were classified into the follow-
ing categories: tinnitus, hearing loss, migraine, phonopho-
bia (i.e., a specific phobia of sound, including pronounced
sound-induced anxiety reactions), tension headache, chronic
daily headache of any type, TMJD, bruxism, osteoarthritis,
any anxiety disorder (including posttraumatic stress disor-
der, obsessivecompulsive disorder, and phonophobia), any
unipolar depressive disorder, any psychiatric disorder (in-
cluding misophonia; Jager et al., 2020; Schröder et al., 2013),
and any functional somatic syndrome (defined as the follow-
ing diagnoses: fibromyalgia, myalgic encephalomyelitis/
chronic fatigue syndrome, chronic daily headache, irritable
bowel syndrome, TMJD, multiple chemical sensitivity,
functional neurological [conversion] disorder, interstitial
cystitis/bladder pain syndrome, and whiplash-associated
disorders). Lastly, we created binary variables to represent
perceived benefit from three classes of medications: benzo-
diazepines, opioids, and gabapentinoids.
Statistical Analyses
In order to determine which aspects of the hyperacusis
phenotype differed meaningfully between individuals classi-
fied as having loudness and pain hyperacusis, we compared
values of each outcome of interest between the two groups
in a Bayesian framework. When the outcome of interest was
categorical (e.g., the presence or absence of tinnitus), we ex-
amined group differences using a Bayesian analogue of the
Pearson chi-squared test (Gûnel & Dickey, 1974; Jamil
et al., 2017). When the outcome of interest was a contin-
uous variable (e.g., age), we examined mean differences
using a Bayesian analogue of the Welch (unequal-variances)
ttest (Kruschke, 2013). Lastly, when the outcome of interest
was ordinal (e.g., a Likert scale item), group differences
were assessed using a Bayesian ordered probit regression
model (Bürkner & Vuorre, 2019; Liddell & Kruschke,
2018), which assumes that a normally distributed continu-
ous latent variable underlies each ordinal dependent vari-
able. Additional details on the specifics of each model are
presented later in this section.
Conducting group comparisons in a Bayesian hypoth-
esis testing framework allowed us to quantify evidence both
for or against a given null hypothesis using Bayes factors
(Kass & Raftery, 1995; Ly et al., 2020; Wagenmakers et al.,
2018). When considering an arbitrary null hypothesis (H
and a mutually exclusive alternative hypothesis (H
), the
Bayes factor (BF
) is defined as the ratio of how likely the
data are under H
divided by how likely the data are under
. Values of BF
greater than 3 are typically considered
to provide substantial evidence for H
over H
, and values
of BF
less than 0.333 are typically considered to provide
substantial evidence for H
over H
(Wagenmakers et al.,
2011). Values of BF
between 0.333 and 3 are typically
considered inconclusive, providing only anecdotalevi-
dence for either H
or H
. Thus, for every group compari-
son, results could be interpreted as (a) providing significant
support for a true group difference on the outcome vari-
able (BF
> 3), (b) providing significant support for the ab-
sence of a group difference on the outcome variable (BF
0.333), or (c) providing inconclusive evidence for or against
a group difference on the outcome variable. With the use
of Bayesian statistics, we are therefore able to characterize
significant differences between groups as well as significant
similarities, allowing for a more nuanced comparison of
loudness and pain hyperacusis relative to more conventional,
frequentist analytic approaches.
All statistical computations were performed in R
(R Core Team, 2020). FN1
Categorical variables were compared
between groups using default GûnelDickey Bayes factors
for contingency tables (as implemented in the BayesFactor
R package; Morey & Rouder, 2018) based on the indepen-
dent multinomial sampling scheme (Gûnel & Dickey, 1974;
Jamil et al., 2017). In this test, the number of individuals in
each clinical group (pain hyperacusis and loudness hypera-
cusis) is treated as fixed, and cell counts are multinomially
distributed within each row of the contingency Table. A
Dirichlet prior with parameters a
= 1 is placed on
the parameters of each multinomial distribution, and the
analytically derived Bayes factor provides evidence for or
against the null hypothesis of equivalent distributions be-
tween groups.
For 2 ×2 contingency tables, the Bayes factor for
the independent multinomial sampling plan reduces to a
test for the equality of two proportions (i.e., the odds ratio
were being compared on binary variables, we additionally
calculated the OR along with its 95% highest-density credi-
ble interval (CrI) using 15,000 Monte Carlo samples from
the joint posterior distribution of the model parameters. For
comparison with a frequentist hypothesis test, group differ-
ences were deemed statistically significantwhen the full
95% CrI excluded OR = 1. Notably, as the null hypothesis
of complete equality between groups is always false at the
population level (Cohen, 1994), we further sought to test
whether the OR was greater than an interval null hypothesis
representing effects that are thought to be too small to be
practically meaningful (Kirk, 1996). This interval, termed
the region of practical equivalence (ROPE; Kruschke &
Liddell, 2018), was defined as the interval OR = [0.905,
1.105] (i.e., values in which log(OR)isbetween0.1 and
0.1). In order to test the null hypothesis that the popula-
tion effect lies within the ROPE (i.e., the difference in
Custom R code to perform statistical analyses can be found on
the ResearchGate profile of the corresponding author (https://www.
rese fil e/Zachary_Wil liams19/pu bli cations). The
remainder of research materials can be obtained from the corresponding
author upon request.
6American Journal of Audiology 118
AJA-20-00209Williams (Author Proof )
proportions in the population is too small to be of practi-
cal importance), we calculated the ROPE Bayes factor
the prior OR distribution falling within the ROPE divided
by the odds of the posterior OR distribution falling within
the ROPE. Like other Bayes factors, BF
allows for the
quantification of evidence for or against the interval null
hypothesis and can be interpreted in the same manner as BF
(with H
being that the true parameter value falls outside of
the ROPE). When BF
provided substantial support in
favor of the parameter value falling within the ROPE, we
deemed the parameters practically equivalent.BF
values were calculated using the R package bayestestR
(Makowski et al., 2019).
When comparing continuous variables between groups,
we utilized a Bayesian ttest similar to that proposed by
Kruschke (2013), albeit without data-dependent prior dis-
tributions. Dependent variables were standardized (M= 0,
SD = 1) and fit to an unequal-variances ttest model, with
a Normal (0, 1) prior on regression coefficients (i.e., the in-
tercept term and mean difference between groups), a Normal
(0, 1) prior on log-transformed standard deviation parameters
for each group, and a Gamma (2, 0.1) prior on ν,thedegrees
of freedom of the t-distribution. Model parameters were
estimated via Markov chain Monte Carlo (MCMC) as
implemented in the R package brms (Bürkner, 2017, 2018).
Posterior distributions of the parameters were based on
40,000 postwarmup MCMC draws from five separate
Markov chains. The primary parameter of interest was
the standardized mean difference between groups (i.e.,
Cohensd), which we summarized using the posterior me-
dian and 95% CrI. Group differences were deemed sta-
tistically significantwhen the full 95% CrI excluded zero.
Tests of the point null hypothesis were also supplemented
with a Bayes factor (BF
) calculated using the Savage
Dickey Density Ratio (Dickey & Lientz, 1970; Wagenmakers
et al., 2010). In order to determine whether dvalues were
large enough to be practically meaningful, we further ex-
amined BF
values to test the interval null hypothe-
sis that the population value of dlies within the interval
[0.1, 0.1].
For ordinal variables, we fit an unequal-variances
ordered probit regression model (Bürkner & Vuorre, 2019;
Liddell & Kruschke, 2018) to the data, which assumes that
the ordinal dependent variable represents a normally dis-
tributed latent variable. In these models, the standard devi-
ation of the reference group (loudness hyperacusis) was
set to 1 in order to set the scale of the latent variable. The
other parameters estimated in the model were similar to
those in the Bayesian ttest (e.g., mean difference, standard
deviation for pain group, Cohensdeffect size), with the
addition of multiple intercept parameters needed to model
the category thresholds of the dependent variable. For these
models, we placed a Normal (0, 1) prior on the mean differ-
ence between groups, a Normal (0, 1) prior on the log-
transformed standard deviation, and a Student-t
(0, 2.5)
prior on each of the threshold values. Model parameters
were estimated in brms using the same MCMC procedure
described for the continuous models. Again, dwas the pri-
mary parameter of interest, and its value was tested against
the point null by examining whether 0 fell within the 95%
CrI (supplemented by BF
calculated using the Savage
Dickey method). Furthermore, practical significance was
determined by examining BF
values based on the in-
terval null hypothesis that the population value of dlies
within [0.1, 0.1].
Availability of Data and Material
Data from the current study are available from the
Coordination of Rare Diseases at Sanford (CoRDS) Regis-
try, organized by Sanford Research. Interested parties can
apply to access the data at https://research.sanfordhealth.
org/rare-disease-registry. Data generated by the authors
of this study from the original CoRDS data (e.g., binary
codes for qualitative responses) can be requested from the
corresponding author if written permission for the author
to share those data is obtained from Sanford Research.
Demographics, Background, and Onset
of Hyperacusis
The study sample was composed of 243 patients with
self-reported hyperacusis (152 pain hyperacusis, 91 loudness
hyperacusis per our operational definitions) between the
ages of 18 and 85 years (M±SD age = 45.65 ± 15.75 years;
49.8% female). The majority of participants were located
within the United States or Canada (n= 160 [66.9%]), and
95% of the sample identified as White. Most participants
(72.6%) had been formally diagnosed with hyperacusis
by a professional (audiologist: 26.9%, otolaryngologist:
33.5%, other physician: 10.1%, other medical professional:
2.2%). More detailed demographic information for the full
sample and loudness/pain subsamples can be found in
T2Table 2.
Comparing demographics between the two groups,
we found that patients with pain hyperacusis were on aver-
age 5.32 (95% CrI [1.16, 9.45]) years younger than patients
with loudness hyperacusis (d=0.345 [0.615, 0.073],
= 4.58). Despite the age difference
between patients with loudness and pain hyperacusis,
the two groups did not meaningfully differ in duration of
hyperacusis symptoms (d=0.144 [0.424, 0.122], BF
0.345, BF
= 0.254). Sex ratio also differed between
the groups, with approximate parity in the pain hyperacu-
sis group and a predominance of women in the loudness
hyperacusis group (pain: 54.5% male, loudness: 36.8% male;
OR = 2.037 [1.193, 3.52], BF
= 5.07, BF
= 2.97). Par-
ticipants in each group were similarly likely to be located in
the United States (pain: 64.0%, loudness: 61.8%; OR = 1.105
[0.639, 1.911], BF
proximately the same proportion of individuals with profes-
sionally diagnosed hyperacusis (pain: 73.0%, loudness: 72.0%;
OR =1.056[0.587,1.955],BF
Williams et al.: Comparison of Loudness and Pain Hyperacusis 7
AJA-20-00209Williams (Author Proof )
Considering the laterality of hyperacusis, the majority
of patients in both groups experienced symptoms bilaterally
(pain: 83.9%, loudness: 78.9%; OR = 1.393 [0.723, 2.739],
= 0.207, BF
= 0.205). However, fewer patients in-
dicated that both ears were affected equally, with practically
equivalent proportions of individuals in each group endors-
ing asymmetric symptoms (pain: 48.0%, loudness: 51.6%;
OR = 0.865 [0.525, 1.487], BF
= 0.193, BF
= 0.103).
Data were not reported regarding the laterality of asymmet-
ric symptoms, although monaural hyperacusis was reported
with approximately equal frequencies in either ear (right:
n= 21, left: n=22).Withregardtocommonhyperacusis
risk factors, the pain hyperacusis group reported signifi-
cantly increased rates of loud noise exposure (pain: 52.6%,
loudness: 39.3%; OR = 1.699 [1.022, 2.948]), although Bayes
factors were inconclusive regarding the practical significance
of this effect (BF
= 1.20, BF
= 0.757). Conversely,
practically equivalent proportions of individuals in each
group reported a history of exposure to traumatic impulse
noise exposures such as blasts and gun fire (pain: 36.4%,
loudness: 31.1%; OR = 1.259 [0.743, 2.226], BF
= 0.221,
= 0.138) and serious head injuries (pain: 25.2%,
loudness: 20.5%; OR = 1.285 [0.681, 2.421], BF
= 0.196,
= 0.155). A diagnosis of hearing loss preceded the
hyperacusis diagnosis in approximately 20% of cases across
groups (pain: 20.3%, loudness: 21.6%; OR = 0.918 [0.484,
1.745], BF
= 0.115), and a relatively
small proportion of individuals in each group reported
wearing hearing aids at the time of the survey (pain: 9.3%,
loudness: 7.2%; OR = 1.255 [0.460, 3.295], BF
= 0.108,
= 0.207). All but two participants reporting hear-
ing aid use (both in the pain group) indicated that they
wore hearing aids binaurally.
QoL and Functional Impairment
Confirmatory factor analysis of the five QoL items in-
dicated that they did not form a unidimensional composite
= 0.913, TLI
= 0.826, RMSEA
AQ4= 0.161;
Savalei, 2020). Thus, scores were compared between groups
on an item-by-item basis using ordered probit models.
Comparing QoL Item 1 (general self-perceived health)
between groups, patients with pain hyperacusis group re-
ported numerically lower scores than patients with loudness
hyperacusis, although scores were practically equivalent be-
tween groups (d=0.132 [0.415, 0.156], BF
= 0.317,
Table 2. Participant demographics and clinical characteristics by hyperacusis subtype.
Demographics and clinical characteristicsAQ3 n
Pain hyperacusis Loudness hyperacusis Full sample
Age (years) 150/89 43.69 (15.58) 48.96 (15.56) 45.65 (15.75)
Sex 145/87
Male 79 (54.5%) 32 (36.8%) 111 (47.8%)
Female 66 (45.5%) 55 (63.2%) 121 (52.2%)
Location 150/89
United States/Canada 102 (68.0%) 58 (65.2%) 160 (66.9%)
Europe 38 (25.3%) 26 (29.%) 64 (26.8%)
Australia/New Zealand 3 (2.0%) 4 (4.5%) 7 (2.9%)
Other 7 (4.7%) 1 (1.1%) 8 (3.3%)
Professional diagnosis 141/86 103 (73.0%) 62 (72.0%) 165 (72.7%)
Duration of hyperacusis 144/87
Less than 1 year 28 (19.4%) 15 (17.2%) 43 (18.6%)
13 years 35 (24.3%) 16 (18.4%) 51 (22.1%)
46 years 28 (19.4%) 13 (14.9%) 41 (17.7%)
710 years 16 (11.1%) 18 (20.7%) 34 (14.7%)
1114 years 9 (6.2%) 7 (8.0%) 16 (6.9%)
15 years or more 28 (19.4%) 18 (19.4%) 46 (19.9%)
Symptom trajectory 147/87
Improved 11 (7.5%) 18 (20.7%) 29 (12.4%)
Same 20 (13.6%) 9 (13.6%) 29 (12.4%)
Worsened 75 (51.0%) 30 (34.5%) 105 (44.9%)
Fluctuated 41 (27.9%) 30 (34.5%) 71 (30.3%)
Symptom laterality 149/90
Bilateral 125 (83.9%) 71 (78.9%) 196 (82.0%)
Unilateral left 11 (7.4%) 11 (12.2%) 22 (9.2%)
Unilateral right 13 (8.7%) 8 (8.9%) 21 (8.8%)
Hearing loss 152/88 73 (48.0%) 42 (47.7%) 115 (47.9%)
Tinnitus 152/91 129 (84.9%) 79 (86.8%) 200 (85.6%)
Functional impairment (010)
Career 142/88 7.67 (2.89) 5.59 (3.40) 6.87 (3.25)
Domestic 147/89 7.15 (2.58) 5.75 (2.79) 6.62 (2.74)
Social 145/89 8.52 (1.82) 7.12 (2.73) 7.99 (2.31)
Note. Values for continuous variables (including 010 rating scales for impairment) are presented as mean (SD), while values for categorical/
ordinal variables are presented as n(%); n
= number of patients in pain/loudness hyperacusis group providing nonmissing response
to question.
8American Journal of Audiology 118
AJA-20-00209Williams (Author Proof )
= 0.231). Similarly, inconclusive results were ob-
served regarding QoL Item 2 (vigorous activity limitation;
d=0.174 [0.463, 0.114], BF
Perhaps unsurprisingly, large group differences were ob-
served in QoL Item 3 (pain interference with enjoyment of
life), with the pain hyperacusis group reporting much lower
QoL due to pain interference (d=1.104 [1.440, 0.772],
). Numerically
lower scores in the pain group were also demonstrated on
QoL Item 4 (tiredness), although results were inconclusive
regarding whether a significant group difference was present
(d=0.205 [0.509, 0.097], BF
On the final QoL item (depression), persons classified as hav-
ing pain hyperacusis again reported lower numerical scores,
but the ROPE Bayes factor indicated practical equivalence
between groups (d=0.175 [0.462, 0.118], BF
= 0.417,
= 0.329).
Three questions on the hyperacusis-specific survey
asked respondents to rate the impact of hyperacusis on
their occupational, domestic, and social functioning on an
11-point scale from 0 (no impact) to 10 (completely ruined
by hyperacusis). Comparing these measures between groups,
individuals with pain hyperacusis reported substantially
more impairment in the occupational domain (d= 0.701
[0.425, 0.987], BF
as well as moderately more impairment in domestic function-
ing (d= 0.527 [0.261, 0.791], BF
= 377, BF
= 161)
and social functioning (d= 0.541 [0.247, 0.840], BF
= 111,
= 70.1).
Symptoms Attributed to Hyperacusis
Pain in the intermittentcategory (e.g., stabbing,
sharp) was by far the most commonly endorsed in both
groups (pain: 63.8%, loudness: 29.3%), with individuals in
the pain hyperacusis group substantially more likely to en-
dorse this symptom (OR = 4.177 [2.328, 7.437], 4.82 ×10
). Pain of the continuoustype (e.g.,
throbbing,”“dull ache)waslessfrequentinbothgroups
(pain: 36.2%, loudness: 19.5%) and similarly more prevalent
in patients with pain hyperacusis (OR =2.272[1.219,4.409],
= 5.07, BF
= 3.84). However, this pattern was
not replicated for the category of neuropathic-likepain,
the prevalence of which was practically equivalent across
the groups (pain: 25.9%, loudness: 18.6%; OR =1.360
[0.661, 2.759], BF
the majority of individuals in both groups reported that
sound-induced pain occurred immediately after hearing the
sound, a sizable minority reported experiencing pain hours
to days after the inciting event (pain: 15.2%, loudness [ex-
cluding individuals who reported no pain]: 36.2%; OR =
0.321 [0.153, 0.682], BF
= 12.5, BF
= 15.1). More-
over, the two groups differed minimally in the duration of
sound-induced pain (d=0.188[0.151, 0.522], BF
= 0.362).
With regard to sound-evoked symptoms, a practically
equivalent proportion of participants in each group en-
dorsed scalp pressure, flu-like symptoms, fluttering/middle
ear myoclonus, hearing changes, sounds such as clicking
or popping, and paresthesia (all BF
< 0.33; see T3Table 3).
Individuals with pain hyperacusis were significantly
more likely to endorse facial pain (pain: 36.9%, loud-
ness: 23.5%; OR = 1.868 [1.027, 3.446], BF
= 1.04), ear fullness (pain: 66.2%, loudness:
51.7%; OR = 1.824 [1.070, 3.131], BF
= 1.79, BF
1.18), and ear pressure (pain: 64.0%, loudness: 47.7%;
OR = 1.933 [1.151, 3.313], BF
= 3.26, BF
= 1.95)
than those with loudness hyperacusis, although Bayes
factors were inconclusive regarding the practical signif-
icance of these differences. Numerically higher propor-
tions of pain hyperacusis patients also endorsed both
jaw pain (pain: 34.0%, loudness: 22.4%; OR = 1.760 [0.950,
3.246], BF
= 0.671) and neck pain
(pain: 34.8%, loudness: 24.7%; OR = 1.601 [0.888, 2.952],
= 0.542, BF
= 0.420), with Bayes factors provid-
ing inconclusive evidence regarding group differences in
symptom endorsement.
Clinical Course and Symptom Fluctuations
The onset of hyperacusis symptoms was described as
sudden in 49.0% of the pain hyperacusis group and 49.4%
of the loudness hyperacusis group, with the difference be-
tween the groups being practically insignificant (OR = 0.982
[0.583, 1.689], BF
= 0.170, BF
= 0.089). When de-
scribing the natural history of hyperacusis symptoms, par-
ticipants categorized their clinical course as Better, Worse,
Same, or Fluctuated. Examining the full 2 ×4 contingency
table (see Table 2), there was substantial evidence for a dif-
ference in symptom trajectories between groups (BF
3.86). Examining each category separately, the pain group
was substantially less likely to endorse symptoms getting
better over time (pain: 7.5%, loudness: 20.7%; OR = 0.320
[0.142, 0.697], BF
= 11.7) and significantly more likely
to endorse symptoms worsening over time (pain: 51.0%,
loudness: 34.5%; OR = 1.96 [1.148, 3.392], BF
= 2.22),
although the ROPE Bayes factor provided inconclusive evi-
dence that the latter difference was large enough to be of
practical significance. Notably, a similar proportion of indi-
viduals in each group indicated that their symptoms had
stayed the same over time (pain: 13.6%, loudness: 10.3%;
OR = 1.316 [0.581, 3.007], BF
= 0.187) or fluctuated
between better and worse (pain: 27.9%, loudness: 34.5%;
OR = 0.734 [0.423, 1.306], BF
= 0.189).
The majority of individuals in both groups reported
experiencing setbacks(defined as specific events that
noticeably worsened symptoms), with only 14.8% of the
loudness group and 13.0% of the pain group stating they
had never experienced temporary symptom exacerbations
(see F1Figure 1). Setbacks were most commonly attributed to
moderate or loud noise exposure, with equivalent proportions
of both groups identifying environmental noises as the cause
(pain: 77.1%, loudness: 70.8%; OR = 1.388 [0.729, 2.763],
= 0.259, BF
= 0.204). A minority of patients
attributed their setbacks to stress, exhaustion, or fatigue,
with inconclusive evidence for differences between groups
Williams et al.: Comparison of Loudness and Pain Hyperacusis 9
AJA-20-00209Williams (Author Proof )
Table 3. Sound-evoked symptom frequencies by hyperacusis subtype.
Symptom n
Pain hyperacusis Loudness hyperacusis OR (95% CrI) BF
Ear pain: intermittent 141/82 90 (63.8%) 24 (29.3%) 4.177 [2.328, 7.437] 4.82 ×10
1.26 ×10
Ear pain: continuous 141/82 51 (36.2%) 16 (19.5%) 2.272 [1.219, 4.409] 5.07 3.84
Ear pain: neuropathic-like 141/82 29 (20.6%) 13 (15.7%) 1.360 [0.661, 2.759] 0.197 0.188
Pain in face 141/85 52 (36.9%) 20 (23.5%) 1.868 [1.027, 3.446] 1.40 1.04
Pain in jaw 141/85 48 (31.6%) 19 (20.9%) 1.760 [0.950, 3.246] 0.875 0.671
Pain in neck 141/85 49 (34.8%) 21 (24.7%) 1.601 [0.888, 2.952] 0.542 0.420
Ear fullness 148/87 98 (66.2%) 45 (51.7%) 1.824 [1.07, 3.131] 1.79 1.18
Ear pressure 150/88 96 (64.0%) 42 (47.7%) 1.933 [1.151, 3.313] 3.26 1.95
Scalp pressure 141/85 45 (31.9%) 26 (30.6%) 1.056 [0.598, 1.902] 0.161 0.098
Flu-like symptoms 141/86 29 (20.6%) 13 (15.1%) 1.417 [0.689, 2.871] 0.218 0.212
Flutter/middle ear myoclonus 152/91 12 (7.9%) 4 (4.4%) 1.69 [0.608, 5.704] 0.136 0.330
Hearing changes/distortion 152/91 7 (4.6%) 7 (7.7%) 0.579 [0.202, 1.634] 0.130 0.364
Sounds (clicking, popping, etc.) 152/91 19 (12.5%) 14 (15.4%) 0.773 [0.374, 1.63] 0.140 0.175
Paresthesia 152/91 9 (5.9%) 4 (4.4%) 1.251 [0.427, 4.317] 0.083 0.236
Reactive tinnitus 122/77 99 (81.1%) 60 (77.9%) 1.217 [0.623, 2.498] 0.170 0.149
Headache exacerbation 122/71 79 (74.8%) 43 (60.6%) 1.204 [0.669, 2.221] 0.212 0.134
Vertigo exacerbation 91/57 46 (50.5%) 57 (40.4%) 1.509 [0.772, 2.917] 0.428 0.264
Note. Odds ratio (OR) and 95% credible interval (CrI) calculated using 15,000 samples from joint posterior distribution of category proportions.
Credible intervals that exclude OR = 1 are presented in bold. Bayes factor (BF) values greater than 3 (providing substantial evidence for a group
difference) are bolded, whereas BF values less than 1/3 (providing substantial evidence against a group difference) are italicized. n
number of patients in pain/loudness hyperacusis group providing codable/nonmissing responses; BF
= Bayes factor testing comparing
proportions between the two groups; BF
= Bayes factor versus the interval null hypothesis log(OR)=[0.1, 0.1], that is, the region of
practical equivalence to 0 (ROPE).
Figure 1. Comparison of setback frequency and severity between patients with loudness and pain hyperacusis. Percentages of individuals in
each group endorsing each response on the Coordination of Rare Diseases at Sanford survey regarding (A) setback frequency and (B) setback
severity. Ordered probit models were used to compare latent mean scores between groups, and the posterior distributions of latent mean scores
are depicted for (C) setback frequency and (B) setback severity. Vertical dotted lines indicate the grand mean score of the sample on each
10 American Journal of Audiology 118
AJA-20-00209Williams (Author Proof )
(pain: 11.0%, loudness: 19.4%; OR = 0.513 [0.232, 1.167],
= 0.469, BF
= 0.631). Notably, individuals in
the pain group reported that they experienced setbacks
significantly more often than individuals in the loudness
group (d= 0.617 [0.345, 0.897], BF
= 672, BF
435). While 62.3% of individuals with pain hyperacusis
reported experiencing setbacks once per month or more,
only 32.1% of individuals in the loudness group reported
having setbacks this frequently (OR = 3.442 [1.939, 6.099],
ing setbacks, those with pain hyperacusis also reported greater
worsening of symptoms during setbacks than those with
loudness hyperacusis (d= 0.616 [0.282, 0.963]), BF
= 137,
= 98.9). However, there was inconclusive evidence
regarding whether the two groups differed in terms of the
amount of time needed to recover from a setback (d= 0.252
[0.138, 0.668], BF
majority of patients (56.1%) reporting that these setbacks
typically last several days. Patients frequently perceived
the need to avoid loud noises during a setback, with the
majority of individuals in both groups responding affirma-
tively to the question, Does the Participants recovery take
longer if they are exposed to a louder noise?(pain: 91.5%,
loudness: 84.7%; OR = 1.922 [0.799, 4.803], BF
= 0.331,
= 0.517).
Comorbid Conditions
As expected, there was high comorbidity between
hyperacusis, tinnitus, and hearing loss in our sample, with
approximately 85% of patients in each group reporting tin-
nitus (pain: 84.9%, loudness: 86.8%; OR = 0.870 [0.410,
1.813], BF
= 0.125, BF
patients reporting some form of hearing loss (pain: 48.0%,
loudness: 47.7%; OR = 1.012 [0.607, 1.713], BF
= 0.166,
= 0.085). Of the patients who endorsed tinnitus,
the proportion whose tinnitus interfered with sleep was
substantially smaller and practically equivalent between
groups (pain: 39.0%, loudness: 50.0%; OR =0.728[0.43,
11.334], BF
= 0.327, BF
= 0.195). A similar propor-
tion of individuals indicated that their tinnitus causes diffi-
culty concentrating while reading in a quiet room, with no
practically meaningful group differences in endorsement
(pain: 60.0%, loudness: 56.4%; OR = 1.166 [0.665, 2.086],
= 0.200, BF
= 0.113). The majority of patients
with comorbid tinnitus also noted that their tinnitus in-
creased in volume with exposure to sounds (pain: 81.1%,
loudness: 77.9%; OR = 1.217 [0.623, 2.498], BF
= 0.170,
= 0.149). Within the group of patients with reac-
tive tinnitus, there was no difference between patients with
loudness hyperacusis and pain hyperacusis in regard to the
duration of tinnitus increases after hearing a loud sound
(d= 0.130 [0.180, 0.448], BF
= 0.312, BF
= 0.229),
with the most common response (40%) being that tinnitus
exacerbations lasted several hours.
Headache disorders were also prevalent in our sample,
with equivalent proportions of individuals in each group
reporting both migraines (pain: 42.8%, loudness: 35.2%;
OR = 1.370 [0.810, 2.357], BF
= 0.316, BF
= 0.197)
and tension headaches (pain: 54.6%, loudness: 48.4%;
OR = 1.280 [0.784, 2.203], BF
= 0.256, BF
= 0.145).
However, chronic daily headaches of any type were experi-
enced significantly more frequently by patients with pain
hyperacusis (pain: 29.4%, loudness: 13.6%; OR = 2.534
[1.281, 5.149], BF
= 6.71, BF
= 6.21). Headaches
occurred frequently in both groups of patients, with indi-
viduals in the pain hyperacusis group experiencing head-
aches slightly more often than individuals in the loudness
hyperacusis group (d= 0.361 [0.086, 0.627], BF
= 5.86,
= 4.17). Average headache severity, rated on a
110 scale, was also higher in the pain group as compared
to the loudness group (pain: M=6.19,loudness:M=5.20;
d= 0.466 [0.175, 0.758], BF
= 29.0, BF
= 18.7).
The majority of patients in both groups who experienced
headaches also endorsed an increased frequency of head-
aches after being exposed to loud noises (pain: 64.8%,
loudness: 60.6%; OR = 1.204 [0.669, 2.221], BF
= 0.212,
= 0.134).
Additional comorbidities reported by over 5% of the
sample included TMJD, bruxism, osteoarthritis, phono-
phobia, anxiety disorders, depressive disorders, psychiatric
disorders in general, and functional somatic syndromes
(see F2Figure 2). Although not reaching the 5% threshold,
it is notable that 10 individuals (4.1%; eight in pain group,
two in loudness group) reported experiencing misophonia
comorbid with their hyperacusis. Phonophobia, including
both self-reported phonophobia diagnoses and descriptions
of pathological anxiety/panic responses evoked by sounds,
was present in just under 10% of the sample and endorsed
to an equivalent degree across groups (pain: 9.2%, loudness:
9.9%; OR = 0.911 [0.391, 2.195], BF
= 0.099, BF
0.165). TMJD was reported by 20% of the sample, with
practically equivalent endorsement across groups (pain:
21.7%, loudness: 17.6%; OR = 1.271 [0.665, 2.498], BF
0.175, BF
= 0.153). Bruxism was also relatively com-
mon, occurring in 35.8% of the sample and similarly across
groups (pain: 35.5%, loudness: 36.3%; OR = 0.965 [0.565,
1.667], BF
= 0.159, BF
= 0.090). Anxiety disorders
(inclusive of phonophobia) were present in 21.8% of the
sample, with a significantly higher rate of endorsement in
the loudness group but inconclusive evidence for a practi-
cally significant difference (pain: 17.1%, loudness: 29.7%;
OR = 0.492 [0.262, 0.893], BF
= 1.73).
Similar patterns were also observed for the categories of
depressive disorders (pain: 10.5%, loudness: 18.7%; OR =
0.515 [0.251, 1.071], BF
any psychiatric disorder (pain: 20.4%, loudness: 34.1%;
OR = 0.501 [0.281, 0.901], BF
= 1.85),
although the group difference in depression prevalence
did not reach statistical significance (i.e., the 95% CrI
included OR = 1). Consistent with previous studies (Paulin
et al., 2016), we found a high prevalence of functional
somatic syndromes in individuals with hyperacusis, with
the pain hyperacusis group numerically more likely to
report one of these conditions (pain: 45.5%, loudness:
34.1%; OR = 1.597 [0.948, 2.749], BF
= 0.724, BF
Williams et al.: Comparison of Loudness and Pain Hyperacusis 11
AJA-20-00209Williams (Author Proof )
0.465). Notably, when excluding chronic daily headache
from the functional somatic syndrome category, nearly
equal proportions of each group endorsed a functional so-
matic syndrome diagnosis (pain: 27.0%, loudness: 26.4%;
OR = 1.023 [0.571, 1.828], BF
= 0.146, BF
= 0.098).
Comorbid Nonauditory Sensory Symptoms
The surveys also included several questions regarding
sensory symptoms in the vestibular, visual, olfactory, and
somatosensory modalities. Vertigo was a commonly reported
symptom, occurring daily or weekly in 27.8% of the sample
(pain: 29.2%, loudness: 25.6%; OR = 1.184 [0.653, 2.139],
= 0.178, BF
= 0.121). Among the patients report-
ing vertigo, neither the severity of vertigo symptoms (d=
0.058 [0.273, 0.394], BF
= 0.260, BF
= 0.176) nor
the frequency of vertigo attacks differed between loudness
and pain groups (d=0.099 [0.382, 0.192], BF
= 0.268,
= 0.181). Additionally, approximately half of pa-
tients experiencing vertigo reported an increase in vertigo
attack frequency or severity after exposure to loud noises
(pain: 50.5%, loudness: 40.4%; OR = 1.509 [0.772, 2.917],
= 0.428, BF
= 0.264). Discomfort due to bright
lights (i.e., photophobia) was also common, occurring daily
or weekly in approximately 40% of patients (pain: 39.7%,
loudness: 39.8%; OR = 0.994 [0.581, 1.756], BF
= 0.168,
= 0.093). Of the patients reporting photophobia,
no differences were found between loudness and pain groups
in terms of photophobia severity (d= 0.016 [0.286, 0.322],
= 0.217, BF
= 0.135) or frequency (d=0.039
[0.248, 0.328], BF
= 0.219, BF
= 0.133). A sizable
proportion of individuals in each group also endorsed be-
ing bothered by strong smells (i.e., osmophobia; pain: 33.6%,
loudness: 38.3%; OR = 0.813 [0.463, 1.432], BF
= 0.213,
= 0.131) and being bothered by touch (pain: 22.4%,
loudness: 23.5%; OR = 0.929 [0.492, 1.747], BF
= 0.146,
= 0.111).
As an exploratory analysis, we also examined the
rates at which the various nonauditory sensory symptoms
co-occurred, additionally investigating the relationships
of these symptoms with psychiatric and functional somatic
syndrome diagnoses. Notably, disturbances in nonauditory
sensory modalities co-occurred at rates substantially higher
than chance. Of the individuals who reported daily or weekly
photophobia, 62.7% endorsed osmophobia (vs. 16% of those
without photophobia; OR = 8.475 [4.508, 16.472]), 35.4%
Figure 2. Comparison of comorbid conditions and nonauditory sensory complaints between patients with loudness and pain hyperacusis.
(A) Percentages of participants with loudness and pain hyperacusis who reported a certain diagnosis/symptom. (B) Posterior densities of each
odds ratio (OR) comparing respondents with pain hyperacusis to those with loudness hyperacusis. OR values are presented with their 95%
CrIs. Thick and thin horizontal lines indicate bounds of the 80% and 95% CrIs, respectively. The gray shaded area indicates the bounds of
the interval null hypothesis (i.e., the region of practical equivalence OR = [0.905, 1.105]). CrI = credible interval; TMJD = temporomandibular joint
12 American Journal of Audiology 118
AJA-20-00209Williams (Author Proof )
endorsed tactile intolerance (vs. 16.3% of those without
photophobia; OR = 2.765 [1.496, 5.44]), and 43.7% en-
dorsed daily or weekly episodes of vertigo (vs. 16.4% of
those without photophobia; OR = 3.871 [2.127, 7.386]).
Similarly, individuals who endorsed osmophobia reported
higher rates of both bothersome tactile experiences (OR =
3.554 [1.884, 7.045]) and daily/weekly vertigo (OR = 3.425
[1.861, 6.501]), and individuals who endorsed tactile intol-
erance had higher rates of daily/weekly vertigo (OR = 4.274
[2.274, 8.561]). Exploratory analyses also revealed that
symptoms in nonauditory sensory modalities were associ-
ated with a functional somatic syndrome diagnosis (photo-
phobia: OR = 2.212 [1.288, 3.865]; tactile intolerance:
OR = 2.719 [1.300, 5.936]; vertigo: OR = 2.755 [1.542,
5.017]), with the exception of osmophobia, the prevalence
of which was equivalent between groups with and with-
out functional somatic syndrome diagnoses (OR =1.403
[0.801, 2.463], BF
= 0.327, BF
= 0.213). Conversely,
the diagnosis of a psychiatric disorder was unrelated to daily/
weekly photophobia (OR = 1.128 [0.613, 2.072], BF
0.202, BF
=0.118)ordaily/weeklyvertigo(OR =
1.207 [0.638, 2.285], BF
= 0.139), al-
though osmophobia was more likely to occur in individuals
with psychiatric diagnoses (OR = 2.807 [1.537, 5.228],
= 44.2, BF
= 19.7). Patients with psychiatric
comorbidities were also numerically more likely to report
tactile intolerance, although evidence for this association
was inconclusive (OR = 1.896 [0.989, 3.617], BF
= 0.927,
= 0.873).
Perceived Response to Hyperacusis Treatments
Of our sample, over 50% of each group had attempted
to treat their hyperacusis with sound therapy, including both
self-directed protocols and those prescribed by professionals
(pain: 59.9%, loudness: 53.8%; OR = 1.273 [0.763, 2.173],
= 0.248, BF
= 0.141). The most common form
of sound therapy was self-administered pink noise (n= 85),
followed by self-administered white noise (n= 84), struc-
tured interventions that included counseling (n= 58), tinni-
tus retraining therapy (n=55),Other(n=48),hearing-aid
sound generators (n= 26), and the Neuromonics protocol
(n= 3). Participants reported engaging in sound therapy
for a median of 12 years, with the loudness hyperacusis
group reporting a moderately longer duration of sound ther-
apy compared to the pain hyperacusis group (d=0.525
[0.908, 0.137], BF
the loudness group also reportedmoreperceivedbenefitfrom
sound therapy than the pain group (d=0.425 [0.806,
0.047], BF
= 2.84). Individuals in the
loudness group were substantially more likely to report
that sound therapy resulted in significant improvementor
hyperacusis [being] almost eliminated(pain: 4.4%, loud-
ness: 22.4%; OR = 0.176 [0.054, 0.557]). However, the pro-
portions of patients reporting (a) no change in symptoms
with sound therapy (pain: 38.5%, loudness: 32.7%; OR =
1.274 [0.618, 2.605], BF
= 0.162) or (b) worsening
tinnitus/hyperacusis (pain: 27.5%, loudness: 18.4%; OR =
1.616 [0.702, 3.801], BF
= 0.313) did not differ mean-
ingfully between the two groups.
The three medication classes reported as beneficial by
eight or more individuals were benzodiazepines (n= 30),
opioids (n= 9), and gabapentinoids (n= 10). Benzodiaze-
pines were reported as beneficial for hyperacusis symptoms
by equivalent proportions of patients in each group (pain:
11.8%, loudness: 13.2%; OR = 0.871 [0.400, 1.870], BF
0.115, BF
= 0.150). In contrast, all individuals report-
ing symptom reduction from opioids or gabapentinoids
were in the pain hyperacusis group, likely due to the spe-
cific effects of these medications on nociceptive processes.
Using a multinational patient registry, the current study
investigated how patients with loudness and pain hyperacusis
differed with respect to demographics, background/onset
characteristics, symptoms, comorbidities, and response to
treatment. On the whole, the two hyperacusis subgroups
were more alike than they were different, yielding equiva-
lent results on the majority of tested variables. However,
the group differences that were detected suggest that pain
hyperacusis is likely a more severe phenotype than loud-
ness hyperacusis, associated with more frequent and severe
setbacks, less improvement over time, a higher burden of
comorbid headache disorders, reduced benefit from sound
therapy, and more overall functional impairment. Notably,
the distinction between loudness and pain hyperacusis was
somewhat blurred in this cohort, as over 90% of patients
with pain hyperacusis perceived everyday sounds as unbear-
ably loud, and over 60% of our loudness hyperacusis
group experienced sound-induced pain at least monthly.
Thus, while our study does not indicate that pain and loud-
ness hyperacusis are pathophysiologically distinct condi-
tions, it does suggest that a pain-predominant phenotype
could serve as a prognostic marker within the hyperacusis
Demographically, the loudness hyperacusis group
was slightly older and more likely to be female, but the
two groups were generally well matched on most other
background variables. However, reported symptom trajecto-
ries differed substantially between the two groups, with the
pain hyperacusis group more likely to report worsening over
time and less likely to report improvement. Individuals with
pain hyperacusis also reported higher scores on all three
functional impairment items when compared with the loud-
ness hyperacusis group. Group differences in QoL items
(other than the pain interference item) were small and
largely not significant, but the directions of these effects
were all consistent with slightly decreased QoL in pain
hyperacusis. Larger studies that implement more reliable
measures of health-related QoL may permit more precise
characterization of differences between these subsets of
the hyperacusis population.
Specific symptoms of hyperacusis were typically re-
ported to a similar degree between the two groups, although
several symptoms such as facial pain, ear pressure, and ear
Williams et al.: Comparison of Loudness and Pain Hyperacusis 13
AJA-20-00209Williams (Author Proof )
fullness were reported slightly more often by patients with
pain versus loudness hyperacusis. However, large group dif-
ferences were found in the domain of setbacks, with the pain
hyperacusis group reporting significantly more frequent and
severe symptom exacerbations than the loudness hyperacusis
group. As the majority of patients in our sample experi-
enced setbacks at least monthly, our study supports the
claims of patient advocates who argue that setbacks should
be considered a core feature of hyperacusis deserving of
additional clinical and research attention (Pollard, 2019).
Future research on this symptom cluster should consider
how to best quantify the frequency, severity, and duration
of setbacks, as such a measure would likely be extremely
useful when evaluating the safety and efficacy of clinical
interventions. Additional research is also necessary to char-
acterize the specific events that trigger setbacks, determine
whether setbacks are associated with measurable changes
in auditory physiology, and investigate the factors that en-
hance or inhibit recovery from setbacks, such as sound
exposure/avoidance and medication use.
Comorbidities were frequent in this sample of hyper-
acusis patients. As expected, tinnitus was the most frequently
reported condition, occurring in 85.6% of our sample (Anari
et al., 1999; Sheldrake et al., 2015). Approximately 80% of
individuals with tinnitus also reported that their tinnitus in-
creased in volume with sound exposure, consistent with previ-
ous reports (Schecklmann et al., 2014). Additional commonly
reported comorbidities included primary headache disorders,
psychiatric disorders, and functional somatic syndromes, all
of which have been associated with hyperacusis in prior stud-
ies (Cederroth et al., 2020; Goebel & Floezinger, 2008; Jüris
et al., 2013; Paulin et al., 2016; Schecklmann et al., 2014).
Most co-occurring conditions were equally prevalent in the
two groups, although individuals with pain hyperacusis
were more likely to report chronic daily headaches. Indi-
viduals with loudness hyperacusis were somewhat more
likely to report comorbid psychiatric disorders, particularly
anxiety disorders, although data were inconclusive regard-
ing the practical significance of this group difference. As
temporal trends in the onset of hyperacusis and these co-
morbid disorders were not assessed in the CoRDS survey,
we were unable to determine whether any of these conditions
could be considered risk markers for developing hyperacusis
(though see Goebel & Floezinger, 2008). Additional research
is necessary to better understand the ways in which comorbid
disorders confer hyperacusis risk and whether hyperacusis it-
self is a risk factor for developing certain comorbidities.
Individuals with both loudness and pain hyperacusis
further reported fairly high rates of nonauditory sensory
complaints, including photophobia, osmophobia, tactile
intolerance, and vertigo attacks, consistent with previous
qualitative findings (Ke et al., 2020). Though the severity
and frequency of nonauditory sensory complaints did not
differ between patients with loudness and pain hyperacusis,
the presence of one nonauditory sensory symptom was
highly predictive of additional symptoms in other modalities.
Moreover, the presence of a functional somatic syndrome
predicted symptoms of photophobia, tactile intolerance,
and vertigo, potentially indicating that these processes may
arise from the pathologic process of central sensitization
thought to underlie functional somatic syndromes (den Boer
et al., 2019; La Touche et al., 2018; Suhnan et al., 2017;
Yunus, 2015). With the exception of osmophobia, sensory
abnormalities were also not related to the presence of psy-
chiatric disorders, indicating that the occurrence and co-
occurrence of sensory symptoms cannot be easily attributed
to the presence of anxiety or trait negative affectivity. Fur-
ther research is warranted to better understand the overlap
between hyperacusis, other sensory complaints, and func-
tional somatic syndromes, and subsequent studies should
seek to determine whether central sensitization delineates a
clinically meaningful hyperacusis subgroup.
Lastly, we investigated relations between hyperacusis
subtypes and response to treatments, including sound ther-
apy and medications. The majority of patients in our sample
had utilized some form of sound therapy to treat hyperacu-
sis, although self-reported treatment outcomes were modest.
One notable finding was that individuals with loudness
hyperacusis were more likely to report substantial improve-
ment with sound therapy than individuals with pain hypera-
cusis, although approximately one third of pain hyperacusis
patients reported at least minor improvement. Additionally,
a substantial minority of patients (27.4% in the pain group,
18.4% in the loudness group) reported symptomatic worsen-
ing of either hyperacusis or tinnitus due to sound therapy,
indicating that these treatments may be harmful to a subset
of the population. The field of hyperacusis research would
benefit substantially from research designed to elucidate fac-
tors that predict differential response to sound therapy in
hyperacusis patients.
Few medications were reported by multiple patients
to be beneficial for symptoms, and only benzodiazepines,
opioids, and gabapentinoids were reported by at least 10% of
respondents. Gabapentinoids and opioids were both only
reported as beneficial in patients with pain hyperacusis,
indicating that certain medications targeting nociceptive
symptoms may be specifically useful for individuals with
pain hyperacusis but not loudness hyperacusis. Notably,
benzodiazepines were the medication class with the most
perceived benefit by patients across hyperacusis groups.
This finding is consistent with the hypothesis that hypera-
cusis is caused by decreased inhibition in the central audi-
tory system (Auerbach et al., 2014; Sheppard et al., 2020)
and indicates that benzodiazepines and other GABAergic
medications may reduce hyperacusis symptoms by counter-
acting maladaptive auditory gain within the central nervous
system. Benzodiazepines have also been found to be benefi-
cial in reducing tinnitus in some cases, although evidence
for this indication is mixed and may not be sufficient to
outweigh the potential harms of these medications (Jufas
& Wood, 2015). However, given the substantial disability
associated with cases of severe hyperacusis and the lack of
established pharmacological treatments, we believe that
additional research should further explore the potential ben-
efits of benzodiazepines in the context of controlled clinical
trials. Specific unanswered questions include (a) whether
14 American Journal of Audiology 118
AJA-20-00209Williams (Author Proof )
benzodiazepines provide long-term symptom reduction
with repeated dosing, (b) whether a short-term course of
benzodiazepines will improve recovery from a setback,
and (c) whether acute doses of benzodiazepines objectively
alter psychoacoustic correlates of hyperacusis such as loud-
ness discomfort levels and loudness growth functions. In
the absence of high-quality randomized controlled trials
supporting the use of benzodiazepines in this population,
we strongly caution against the routine prescribing of ben-
zodiazepines for patients with hyperacusis, although judi-
cious use of these medications (with frequent re-assessment
of benefits and harms) may be warranted in some cases.
Given the promising open-label results seen with safer medi-
cations such as nortriptyline and topiramate (Abouzari et al.,
2020), clinicians should consider prescribing benzodiazepines
only in cases that fail to respond to more conservative phar-
macotherapies and/or sound therapy.
Despite the rich information gleaned from this investi-
gation, our study is not without limitations. Most notably,
all information was self-reported by patients in the CoRDS
Registry and was unable to be empirically verified. More-
over, participants in this registry were self-selected and quite
possibly not representative of the broader hyperacusis popu-
lation. In particular, it may be the case that individuals
with pain hyperacusis were overrepresented in this sample, as
noise-induced pain is a particularly strong focus of Hypera-
cusis Research Ltd., the organization sponsoring the
CoRDS hyperacusis survey. Given that many individuals
in this registry were recruited from social media hyperacu-
sis support groups, it is also possible that individuals in
our sample differed from the population of hyperacusis
patients in terms of chronicity of symptoms, comorbid
conditions, illness severity, or sociodemographic factors
(Sautier et al., 2014). Another limitation concerns our
operationalization of pain hyperacusis. As no clinical or
research criteria have been established to parse hyperacu-
sis into loudness and pain subtypes, we chose the arbitrary
criterion of pain experienced daily based on (a) preliminary
results from this sample (Pollard, 2019) and (b) the notion
that pain caused by everyday soundswould result in
pain experienced at least daily. However, it may poten-
tially be the case that individuals who report no sound-
induced pain whatsoever differ systematically from those
who report sound-induced pain in any capacity, and addi-
tional research will be needed to determine whether such
subgroups provide additional prognostic information. A
third limitation is the lack of data in the CoRDS registry
on the perceived efficacy of cognitive-behavioral therapy
(CBT; Aazh et al., 2019), which has demonstrated some
efficacy in reducing hyperacusis symptoms and improving
loudness discomfort levels in a single randomized controlled
trial (Jüris et al., 2014). Future research investigating the
utility of CBT as a potential treatment for hyperacusis
should specifically investigate the degree to which response
to CBT may differ between individuals with and without
pain hyperacusis. Notably, as CBT has demonstrated effi-
cacy in treating other forms of chronic pain with diverse
etiologies (Hoffman et al., 2007; Morley et al., 1999; Pike
et al., 2016), the adaptation of existing CBT protocols to
pain hyperacusis remains a potentially fruitful area of fu-
ture research. Lastly, the information in the CoRDS Reg-
istry was gathered using an ad hoc survey measure rather
than previously validated questionnaires. As a result, it is
difficult to compare this sample of patients to other groups in
terms of hyperacusis or tinnitus severity, comorbid anxiety/
depression symptoms, QoL, and other constructs of inter-
est. Future survey studies in this population should addition-
ally include validated measures of sound tolerance symptoms
(Fackrell & Hoare, 2018; Greenberg & Carlos, 2018) and
other patient-reported outcomes (Cella et al., 2007). De-
spite the limitations of the CoRDS Registry as a source of
information on the phenotype of hyperacusis, we believe
that the information gathered from CoRDS and other pa-
tient surveys has the potential to identify important aspects
of the hyperacusis experience from the perspective of pa-
tients and to guide clinicians and researchers in providing
higher quality care to this particular patient population.
Author Contributions
Zachary J. Williams: Conceptualization (Lead), Data
curation (Lead), Formal analysis (Lead), Investigation (Lead),
Methodology (Lead), Visualization (Lead), Writing original
draft (Lead), Writing review & editing (Lead). Evan
Suzman: Conceptualization (Supporting), Methodology
(Supporting), Visualization (Supporting), Writing original
draft (Supporting), Writing review & editing (Equal).
Tiffany G. Woynaroski: Conceptualization (Supporting),
Formal analysis (Supporting), Resources (Lead), Supervi-
sion (Lead), Writing review & editing (Equal).
This work was supported by National Institute on Deafness
and Other Communication Disorders Grant F30-DC019510
(awarded to Z. J. W.), National Institute of General Medical Sci-
ences Grant T32-GM007347 (awarded to Z. J. W.), and the Nancy
Lurie Marks family foundation (awarded to Z. J. W./T. G. W.). The
authors would like to acknowledge Hyperacusis Research, Ltd. and
Sanford Research for developing and maintaining the CoRDS hyper-
acusis registry, as well as Alyssa Mendel of Sanford Research for
assisting in data curation and management. The authors would
also like to thank Maura Black, David Treworgy, and other hyper-
acusis patient-advocates for their hard work in recruiting individ-
uals experiencing hyperacusis to register with CoRDS.
Aazh, H., Landgrebe, M., Danesh, A. A., & Moore, B. C. J. (2019).
Cognitive behavioral therapy for alleviating the distress caused
by tinnitus, hyperacusis and misophonia: Current perspectives.
Psychology Research and Behavior Management, 12, 9911002.
Abouzari, M., Tan, D., Sarna, B., Ghavami, Y., Goshtasbi, K.,
Parker, E. M., Lin, H. W., & Djalilian, H. R. (2020). Efficacy
of multi-modal migraine prophylaxis therapy on hyperacusis
patients. Annals of Otology, Rhinology & Laryngology, 129(5),
Williams et al.: Comparison of Loudness and Pain Hyperacusis 15
AJA-20-00209Williams (Author Proof )
Acquadro, C., Berzon, R., Dubois, D., Leidy, N. K., Marquis, P.,
Revicki, D., & Rothman, M. (2003). Incorporating the patients
perspective into drug development and communication: An ad
hoc task force report of the patient-reported outcomes (PRO)
harmonization group meeting at the food and drug administra-
tion. Value in Health, 6(5), 522531.
Anari, M., Axelsson, A., Eliasson, A., & Magnusson, L. (1999). Hy-
persensitivity to sound: Questionnaire data, audiometry and
classification. Scandinavian Audiology, 28(4), 219230. https://
Assi, H., Moore, R. D., Ellemberg, D., & Hébert, S. (2018). Sensi-
tivity to sounds in sport-related concussed athletes: A new clin-
ical presentation of hyperacusis. Scientific Reports, 8(1), 9921.
Auerbach, B. D. (2019, January, 1). Physiological mechanisms of
hyperacusis: An update. ENT & Audiology News, 27(6). https://
Auerbach, B. D., Rodrigues, P. V., & Salvi, R. J. (2014). Central
gain control in tinnitus and hyperacusis. Frontiers in Neuro-
logy, 5(6), 206.
Baguley, D. M., & Hoare, D. J. (2018). Hyperacusis: Major re-
search questions. HNO, 66(5), 358363.
Bennett, R. M., Jones, J., Turk, D. C., Russell, I. J., & Matallana, L.
(2007). An internet survey of 2,596 people with fibromyalgia.
BMC Musculoskeletal Disorders, 8(1), 27.
Brandy, W. T., & Lynn, J. M. (1995). Audiologic findings in hyper-
acusic and nonhyperacusic subjects. American Journal of Audi-
ology, 4(1), 4651.
Bürkner, P.-C. (2017). brms: An R package for Bayesian multilevel
models using Stan. Journal of Statistical Software, 80(1). https://
Bürkner, P.-C. (2018). Advanced Bayesian multilevel modeling
with the R package brms. The R Journal, 10(1), 395411. https://
Bürkner, P.-C., & Vuorre, M. (2019). Ordinal regression models
in psychology: A tutorial. Advances in Methods and Practices
in Psychological Science, 2(1), 77101.
Cederroth, C. R., Lugo, A., Edvall, N. K., Lazar, A., Lopez-Escamez,
J.-A., Bulla, J., Uhlen, I., Hoare,D.J.,Baguley,D.M.,Canlon,
B., & Gallus, S. (2020). Association between hyperacusis and
tinnitus. Journal of Clinical Medicine, 9(8), 2412. https://doi.
Cella, D., Yount, S., Rothrock, N., Gershon, R., Cook, K., Reeve, B.,
Ader, D., Fries, J. F., Bruce, B., Rose, M., & PROMIS Coopera-
tive Group. (2007). The patient-reported outcomes measurement
information system (PROMIS): Progress of an NIH roadmap
cooperative group during its first two years. Medical Care,
45(5 Suppl. 1),,S3S11.
Chu, L., Valencia, I. J., Garvert, D. W., & Montoya, J. G.
(2018). Deconstructing post-exertional malaise in myalgic
encephalomyelitis/ chronic fatigue syndrome: A patient-centered,
cross-sectional survey. PLOS ONE, 13(6), Article e0197811.
Cohen, J. (1994). The earth is round (p< .05). American Psychol-
ogist, 49(12), 9971003.
den Boer, C., Dries, L., Terluin, B., van der Wouden, J. C.,
Blankenstein, A. H., van Wilgen, C. P., Lucassen, P., &
van der Horst, H. E. (2019). Central sensitization in chronic
pain and medically unexplained symptom research: A system-
atic review of definitions, operationalizations and measure-
ment instruments. Journal of Psychosomatic Research, 117,
Dickey, J. M., & Lientz, B. P. (1970). The weighted likelihood
ratio, sharp hypotheses about chances, the order of a Markov
chain. The Annals of Mathematical Statistics, 41(1), 214226.
Dworkin, R. H., Turk, D. C., Revicki, D. A., Harding, G., Coyne,
K. S., Peirce-Sandner, S., Bhagwat, D., Everton, D., Burke, L. B.,
Cowan, P., Farrar, J. T., Hertz, S., Max, M. B., Rappaport, B. A.,
&Melzack,R.(2009). Development and initial validation of an
expanded and revised version of the Short-form McGill Pain
Questionnaire (SF-MPQ-2). PAIN®, 144(1), 3542. https://doi.
Fackrell, K., & Hoare, D. J. (2018). Scales and questionnaires
for decreased sound tolerance. In M. A. Fagelson & D. M.
Baguley (Eds.), Hyperacusis and disorders of sound intolerance
(pp. 4358). Plural.
Fackrell, K., Stratmann, L., Kennedy, V., MacDonald, C., Hodgson,
H., Wray, N., Farrell, C., Meadows, M., Sheldrake, J., Byrom,
P., Baguley, D. M., Kentish, R., Chapman, S., Marriage, J.,
Phillips, J., Pollard, T., Henshaw, H., Gronlund, T. A., & Hoare,
D. J. (2019). Identifying and prioritising unanswered research
questions for people with hyperacusis: James Lind Alliance
Hyperacusis Priority Setting Partnership. BMJ Open, 9(11),
Article e032178.
Fagelson, M. A., & Baguley, D. M. (2018). Disorders of sound tol-
erance: History and terminology. In M. A. Fagelson & D. M.
Baguley (Eds.), Hyperacusis and disorders of sound intolerance:
Clinical and research perspectives (pp. 314). Plural.
Flores, E. N., Duggan, A., Madathany, T., Hogan, A. K., Márquez,
F. G., Kumar, G., Seal, R. P., Edwards, R. H., Liberman, M. C.,
&García-Añoveros,J.(2015). A non-canonical pathway from
cochlea to brain signals tissue-damaging noise. Current Biol-
ogy, 25(5), 606612.
Goebel, G., & Floezinger, U. (2008). Pilot study to evaluate psy-
chiatric co-morbidity in tinnitus patients with and without
hyperacusis. Audiological Medicine, 6(1), 7884. https://doi.
Greenberg, B., & Carlos, M. (2018). Psychometric properties and
factor structure of a new scale to measure hyperacusis: Introduc-
ing the inventory of hyperacusis symptoms. Ear and Hearing,
39(5), 10251034.
Gûnel, E., & Dickey, J. (1974). Bayes factors for independence in
contingency tables. Biometrika, 61(3), 545557.
Hébert, S., Fournier, P., & Noreña, A. (2013). The auditory sensi-
tivity is increased in tinnitus ears. Journal of Neuroscience,
33(6), 23562364.
Hoffman, B. M., Papas, R. K., Chatkoff, D. K., & Kerns, R. D.
(2007). Meta-analysis of psychological interventions for chronic
low back pain. Health Psychology, 26(1), 19.
Jager, I., de Koning, P., Bost, T., Denys, D., & Vulink, N. (2020).
Misophonia: Phenomenology, comorbidity and demographics
in a large sample. PLOS ONE, 15(4), Article e0231390. https://
Jamil, T., Ly, A., Morey, R. D., Love, J., Marsman, M., &
Wagenmakers, E.-J. (2017). Default Gunel and Dickey
Bayes factors for contingency tables. Behavior Research Methods,
49(2), 638652.
16 American Journal of Audiology 118
AJA-20-00209Williams (Author Proof )
Jastreboff, P. J., & Jastreboff, M. M. (2015). Decreased sound tol-
erance: Hyperacusis, misophonia, diplacousis, and polyacousis.
In M. J. Aminoff, F. Boller, & D. F. Swaab (Eds.), Handbook
of clinical neurology (Vol. 129, pp. 375387). Elsevier. https://
Jufas, N. E., & Wood, R. (2015). The use of benzodiazepines for tin-
nitus: Systematic review. The Journal of Laryngology & Otology,
129(S3), S14S22.
Jüris, L., Andersson, G., Larsen, H. C., & Ekselius, L. (2013). Psychi-
atric comorbidity and personality traits in patients with hyperacu-
sis. International Journal of Audiology, 52(4), 230235. https://
Jüris, L., Andersson, G., Larsen, H. C., & Ekselius, L. (2014).
Cognitive behaviour therapy for hyperacusis: A randomized
controlled trial. Behaviour Research and Therapy, 54, 3037.
Kanazawa, M., Miwa, H., Nakagawa, A., Kosako, M., Akiho, H.,
& Fukudo, S. (2016). Abdominal bloating is the most bother-
some symptom in irritable bowel syndrome with constipation
(IBS-C): A large population-based Internet survey in Japan.
BioPsychoSocial Medicine, 10(1), 19.
Kass, R. E., & Raftery, A. E. (1995). Bayes factors. Journal of the
American Statistical Association, 90(430), 773795. https://doi.
Ke, J., Du, Y., Tyler, R. S., Perreau, A., & Mancini, P. C. (2020).
Complaints of people with hyperacusis. Journal of the American
Academy of Audiology, 31(8), 553558.
Kirk, R. E. (1996). Practical significance: A concept whose time
has come. Educational and Psychological Measurement, 56(5),
Kruschke, J. K. (2013). Bayesian estimation supersedes the t test.
Journal of Experimental Psychology: General, 142(2), 573603.
Kruschke, J. K., & Liddell, T. M. (2018). The Bayesian new statis-
tics: Hypothesis testing, estimation, meta-analysis, and power
analysis from a Bayesian perspective. Psychonomic Bulletin &
Review, 25(1), 178206.
La Touche, R., Paris-Alemany, A., Hidalgo-Pérez, A., López-de-
Uralde-Villanueva, I., Angulo-Diaz-Parreño, S., & Muñoz-
García, D. (2018). Evidence for central sensitization in patients
with temporomandibular disorders: A systematic review and
meta-analysis of observational studies. Pain Practice, 18(3),
Liddell, T. M., & Kruschke, J. K. (2018). Analyzing ordinal data
with metric models: What could possibly go wrong? Journal of
Experimental Social Psychology, 79, 328348.
Liu, C., Glowatzki, E., & Fuchs, P. A. (2015). Unmyelinated type
II afferent neurons report cochlear damage. Proceedings of the
National Academy of Sciences, 112(47), 1472314727. https://
Ly, A., Stefan, A., van Doorn, J., Dablander, F., van den Bergh,
D., Sarafoglou, A., Kucharský, S., Derks, K., Gronau, Q. F.,
Raj, A., Boehm, U., van Kesteren, E.-J., Hinne, M., Matzke,
D., Marsman, M., & Wagenmakers, E.-J. (2020). The Bayesian
methodology of Sir Harold Jeffreys as a practical alternative
to the p value hypothesis test. Computational Brain & Behav-
ior, 3(2), 153161.
Makowski, D., Ben-Shachar, M. S., Chen, S. H. A., & Lüdecke, D.
(2019). Indices of effect existence and significance in the Bayesian
framework. Frontiers in Psychology, 10.
Makowski, D., Ben-Shachar, M. S., & Lüdecke, D. (2019). BayestestR:
Describing effects and their uncertainty, existence and signifi-
cance within the Bayesian framework. Journal of Open Source
Software, 4(40), 1541.
Morey, R. D., & Rouder, J. N. (2018). BayesFactor: Computation
of bayes factors for common designs. https://CRAN.R-project.
Morley, S., Eccleston, C., & Williams, A. (1999). Systematic re-
view and meta-analysis of randomized controlled trials of cog-
nitive behaviour therapy and behaviour therapy for chronic
pain in adults, excluding headache. Pain, 80(1), 113. https://
Noreña, A. J., & Chéry-Croze, S. (2007). Enriched acoustic envi-
ronment rescales auditory sensitivity. NeuroReport, 18(12),
Noreña, A. J., Fournier, P., Londero, A., Ponsot, D., & Charpentier, N.
(2018). An integrative model accounting for the symptom clus-
ter triggered after an acoustic shock. Trends in Hearing, 22,
Paulin, J., Andersson, L., & Nordin, S. (2016). Characteristics of
hyperacusis in the general population. Noise & Health, 18(83),
Phillips, D. P., & Carr, M. M. (1998). Disturbances of loudness
perception. Journal of the American Academy of Audiology,
9(5), 371379.
Pienkowski, M., Tyler, R. S., Roncancio, E. R., Jun, H. J., Brozoski,
T., Dauman, N., Coelho, C. B., Andersson, G., Keiner, A. J.,
Cacace, A. T., Martin, N., & Moore, B. C. J. (2014). A review
of hyperacusis and future directions: Part II. Measurement,
mechanisms, and treatment. American Journal of Audiology,
23(4), 420436.
Pike, A., Hearn, L., & de C Williams, A. C. (2016). Effectiveness
of psychological interventions for chronic pain on health
care use and work absence: Systematic review and meta-
analysis. Pain, 157(4), 777785.
Pollard, B. (2019). Clinical advancements for managing hyperacu-
sis with pain. The Hearing Journal, 72(10), 10.
R Core Team. (2020). R: A language and environment for statisti-
cal computing (4.0.2) [Computer software]. R Foundation for
Statistical Computing.
Rouw, R., & Erfanian, M. (2018). A large-scale study of misopho-
nia. Journal of Clinical Psychology, 74(3), 453479. https://doi.
Sautier, L., Mehnert, A., Höcker, A., & Schilling, G. (2014). Par-
ticipation in patient support groups among cancer survivors:
Do psychosocial and medical factors have an impact? Euro-
pean Journal of Cancer Care, 23(1), 140148.
Savalei, V. (2020). Improving fit indices in structural equation
modeling with categorical data. Multivariate Behavioral Re-
Schecklmann, M., Landgrebe, M., Langguth, B., & the TRI Data-
base Study Group. (2014). Phenotypic characteristics of hyper-
acusis in tinnitus. PLOS ONE, 9(1), Article e86944. https://
Schröder, A. E., Vulink, N., & Denys, D. (2013). Misophonia:
Diagnostic criteria for a new psychiatric disorder. PLOS
ONE, 8(1), Article e54706.
Sheldrake, J., Diehl, P. U., & Schaette, R. (2015). Audiometric
characteristics of hyperacusis patients. Frontiers in Neurology,
6(1), 105.
Williams et al.: Comparison of Loudness and Pain Hyperacusis 17
AJA-20-00209Williams (Author Proof )
Sheppard, A., Stocking, C., Ralli, M., & Salvi, R. (2020). A review
of auditory gain, low-level noise and sound therapy for tinnitus
and hyperacusis. International Journal of Audiology, 59(1), 515.
Suhnan, A. P., Finch, P. M., & Drummond, P. D. (2017). Hypera-
cusis in chronic pain: Neural interactions between the auditory
and nociceptive systems. International Journal of Audiology,
56(11), 801809.
Trudeau, S. (2013). CoRDS registry: An HIT case study concerning
setup and maintenance of a disease registry. In S. Sarnikar, D.
Bennett, & M. Gaynor (Eds.), Cases on Healthcare Information
Technology for Patient Care Management (pp. 197207). IGI
Tyler, R. S., Pienkowski, M., Roncancio, E. R., Jun, H. J.,
Brozoski, T., Dauman, N., Coelho, C. B., Andersson, G.,
Keiner, A. J., Cacace, A. T., Martin, N., & Moore, B. C. J.
(2014). A review of hyperacusis and future directions: Part I.
Definitions and manifestations. American Journal of Audiol-
ogy, 23(4), 402419.
Wagenmakers, E.-J., Lodewyckx, T., Kuriyal, H., & Grasman, R.
(2010). Bayesian hypothesis testing for psychologists: A tutorial
on the SavageDickey method. Cognitive Psychology, 60(3),
Wagenmakers, E.-J., Marsman, M., Jamil, T., Ly, A., Verhagen, J.,
Love, J., Selker, R., Gronau, Q. F., Šmíra, M., Epskamp, S.,
Matzke, D., Rouder, J. N., & Morey, R. D. (2018). Bayesian
inference for psychology. Part I: Theoretical advantages and
practical ramifications. Psychonomic Bulletin & Review, 25(1),
Wagenmakers, E.-J., Wetzels, R., Borsboom, D., & van der Maas,
H. L. J. (2011). Why psychologists must change the way they
analyze their data: The case of psi: Comment on Bem (2011).
Journal of Personality and Social Psychology, 100(3), 426432.
Weldring, T., & Smith, S. M. S. (2013). Article commentary: Patient-
reported outcomes (PROs) and patient-reported outcome mea-
sures (PROMs). Health Services Insights, 6, HSI.S11093. https://
Williams, Z. J., He, J. L., Cascio, C. J., & Woynaroski, T. G.
(2021). A review of decreased sound tolerance in autism: Defi-
nitions, phenomenology, and potential mechanisms. Neuro-
science & Biobehavioral Reviews, 121, 117.
Williams, Z. J., Suzman, E., & Woynaroski, T. G. (2021). Preva-
lence of decreased sound tolerance (hyperacusis) in individuals
with autism spectrum disorder: A meta-analysis. Ear and Hear-
ing. Advance online publication.
Wu, J. S., Vyas, P., Glowatzki, E., & Fuchs, P. A. (2018). Op-
pos1ing expression gradients of calcitonin-related polypeptide
alpha (Calca/Cgrpα) and tyrosine hydroxylase (Th) in type II
afferent neurons of the mouse cochlea. Journal of Compa-
rative Neurology, 526(3), 425438.
Yunus, M. B. (2015). Editorial review (Thematic issue: An update
on central sensitivity syndromes and the issues of nosology
and psychobiology). Current Rheumatology Reviews, 11(2),
Zelaya, C. E., Lucas, J. W., & Hoffman, H. J. (2015). Percentage
of adults with selected hearing problems, by type of problem
and age groupNational Health Interview Survey, United
States, 2014. Morbidity and Mortality Weekly Report, 64(37),
18 American Journal of Audiology 118
AJA-20-00209Williams (Author Proof )
... Relationships between hyperacusis-like symptoms and occupational noise exposure have been documented in teachers and childcare workers (Anari et al., 1999;Fredriksson et al., 2021;J€ uris et al., 2013;Meuer and Hiller, 2015;Sj€ odin et al., 2012), military personnel (Muhr and Rosenhall, 2010), and call center operators (McFerran and Baguley, 2007;Parker et al., 2014). Complaints of hyperacusis are also common amongst musicians (Anari et al., 1999;Couth et al., 2020;Halevi-Katz et al., 2015;K€ ah€ ari et al., 2003;Laitinen and Poulsen, 2008;Liberman et al., 2016;Schmuziger et al., 2006;Toppila et al., 2011;Wartinger et al., 2019) and individuals with high levels of self-reported noise exposure (Smit et al., 2021;Smith et al., 2019;Williams et al., 2021). ...
... In addition to heightened emotional reactions to sound, most patients with hyperacusis report some form of soundinduced pain, with 52.6% indicating that their pain symptoms were triggered by exposure to loud noise (Williams et al., 2021). Despite the prevalence of pain hyperacusis, the mechanisms underlying sound-evoked pain remain elusive. ...
Full-text available
Hyperacusis is a recognized perceptual consequence of acoustic overexposure that can lead to debilitating psychosocial effects. Despite the profound impact of hyperacusis on quality of life, clinicians and researchers lack objective biomarkers and standardized protocols for its assessment. Outcomes of conventional audiologic tests are highly variable in the hyperacusis population and do not adequately capture the multifaceted nature of the condition on an individual level. This presents challenges for the differential diagnosis of hyperacusis, its clinical surveillance, and evaluation of new treatment options. Multiple behavioral and objective assays are emerging as contenders for inclusion in hyperacusis assessment protocols but most still await rigorous validation. There remains a pressing need to develop tools to quantify common nonauditory symptoms, including annoyance, fear, and pain. This review describes the current literature on clinical and investigational tools that have been used to diagnose and monitor hyperacusis, as well as those that hold promise for inclusion in future trials.
... Personal accounts of autism often describe certain sounds as physically painful (Elwin et al., 2012;Howe & Stagg, 2016 (Dunlop et al., 2016;Khalfa et al., 2004;Ohmura et al., 2019), confirming the presence of loudness hyperacusis in those participants. Notably, an unpublished study by our group of 243 patients with self-reported hyperacusis contained four individuals on the autism spectrum (Williams, Suzman, et al., 2020a), two of whom reported pain hyperacusis (one with comorbid misophonia and the other with comorbid phonophobia) and two of whom reported loudness hyperacusis (one with comorbid phonophobia and the other reporting no other DST conditions). Three of these individuals (two loudness, one pain) had also received diagnoses of hyperacusis from physicians or audiologists. ...
Full-text available
Atypical behavioral responses to environmental sounds are common in autistic children and adults, with 50-70% of this population exhibiting decreased sound tolerance (DST) at some point in their lives. This symptom is a source of significant distress and impairment across the lifespan, contributing to anxiety, challenging behaviors, reduced community participation, and school/workplace difficulties. However, relatively little is known about its phenomenology or neurocognitive underpinnings. The present article synthesizes a large body of literature on the phenomenology and pathophysiology of DST-related conditions to generate a comprehensive theoretical account of DST in autism. Notably, we argue against conceptualizing DST as a unified construct, suggesting that it be separated into three phenomenologically distinct conditions: hyperacusis (the perception of everyday sounds as excessively loud or painful), misophonia (an acquired aversive reaction to specific sounds), and phonophobia (a specific phobia of sound), each responsible for a portion of observed DST behaviors. We further elaborate our framework by proposing preliminary neurocognitive models of hyperacusis, misophonia, and phonophobia that incorporate neurophysiologic findings from studies of autism.
Full-text available
**See the interactive i-poster at the ASHA 2022 website (** Abstract: Introduction: Hyperacusis is a form of decreased sound tolerance (DST) in which sound of moderate intensity is perceived as excessively loud, painful, and/or overwhelming (Fackrell et al., 2019). This condition is frequently observed in autistic children and adults, with an estimated lifetime prevalence of 50–70% in this population (Williams et al., 2021a). Notably, few studies have systematically examined the clinical manifestations of autism-associated hyperacusis, and it remains unclear whether hyperacusis presents differently in autistic and non-autistic people. The present study seeks to address this gap in the literature by systematically examining the clinical features of hyperacusis in a multinational sample of autistic and non-autistic adults. Methods: Participants were drawn from an online sample of 519 adults (aged 18–72 years) who identified as sensitive to sound and endorsed either (a) a prior diagnosis of a DST condition (hyperacusis, misophonia, or phonophobia), (b) previously seeking medical care specifically for DST symptoms, or (c) current DST symptoms that result in clinically significant disability. All participants completed a series of self-report questionnaires measuring demographic characteristics, medical history, multiple types of DST symptomatology, tinnitus, psychopathology, somatic symptoms, and quality of life. From this sample, the current study examined the subset of participants meeting operational criteria for hyperacusis who self-reported either a prior professional diagnosis of autism (n=95; M±SD age 30.95±9.88 years; 42.1% women, 45.3% non-binary gender) or no prior diagnosis of autism (n=80; M±SD age 37.39±13.51 years; 52.5% women, 8.8% non-binary gender). Individuals with hyperacusis who identified as autistic but did not have a formal diagnosis (n=85) were excluded from the present analyses. The clinical manifestations of hyperacusis were then compared in the autistic (AUT) and non-autistic (NON) groups using Bayesian estimation. Results: Participants in both groups reported that they often perceived everyday sounds as excessively or uncomfortable loud (median frequency 1–3 times per day), with no appreciable difference between groups (d=0.138, CrI95% [-0.181,0.446]). Sound-induced pain in one or both ears was also common (median frequency 4–7 times per week), although it occurred slightly less frequently in the AUT group (d=-0.309 [-0.624,-0.004]). Pain hyperacusis, defined as hyperacusis with noise-induced pain on more days than not, was reported by 45.3% of the AUT group and 62.5% of the NON group (OR=0.503 [0.274,0.904]). Degree of disability due to hyperacusis was also moderately lower in autistic adults (d=-0.500 [0.169,0.817]), although the autistic adults also reported using ear protection for more hours per day (d=0.563 [0.224, 0.898]). Autistic adults were also substantially more likely to endorse being able to tolerate pleasant sounds (e.g., one’s preferred music) at high levels without experiencing pain or discomfort (AUT: 56.8%, NON: 25.0%; OR=3.86 [2.04,7.36]). The onset of hyperacusis symptoms was much earlier in the AUT group than the NON group (d=1.46 [1.09,1.84]), with 61.1% of autistic adults endorsing hyperacusis as a lifelong problem and 87.4% reporting symptom onset before age 11 (compared to 15.0% and 23.8% of non-autistic adults). Very few autistic adults (5.3%) identified a specific event or trauma preceding the onset of their hyperacusis symptoms, although this was the case for nearly half of the NON group (47.5%). Comorbidities such as hearing loss and tinnitus were more prevalent in the NON group, whereas autistic adults reported more non-autism psychiatric diagnoses, higher levels of anxiety and perceived stress, more somatic symptoms, and more difficulty tolerating non-auditory sensory stimuli. Of the 86.9% of the sample who endorsed any noise-induced pain, over 95% endorsed immediate pain (occurring at the same time as the painful sound), and 67% endorsed some form of delayed pain that occurred after the painful sound stopped. In almost all cases, immediate sound-induced pain was localized to the ear or periauricular region, and in approximately 80% of cases, the character of this pain was described as “sharp” or “stabbing.” Groups did not differ in terms of immediate pain intensity or duration, although the non-autistic group reported that immediate pain was more frequently accompanied by delayed pain of a different character (d=-0.443 [-0.794,-0.091]). Most participants with delayed pain also localized this pain to the ear or periauricular region (AUT: 71.7%, NON: 95.8%). The character of delayed pain was most frequently described as aching and/or throbbing (AUT: 88.7%, NON: 77.1%). Though the intensity of delayed pain did not differ between groups, the autistic group reported significantly less time between the pain-inducing sound and the average onset of pain (d=-0.754 [-1.21,-0.308]) and a shorter duration of delayed pain (d=-0.648 [-1.08,-0.223]). Hyperacusis symptom flare-ups (“setbacks”) lasting several hours or longer were also common in both groups (past year prevalence: AUT: 68.4%, NON: 65.0%), with a median frequency of 2–4 episodes per month over the past year. Though loud noise was the most common perceived cause of symptom flare-ups in both groups, a smaller proportion of symptom flare-ups in the autistic group were attributed to specific noises (d=-0.690 [-1.09, -0.301]). Autistic adults also reported a shorter average duration of flare-ups (d=-0.385 [-0.762, -0.007]) and less flare-up related impairment (d=-0.470 [-0.861, -0.087]). Notably, the frequency of certain symptoms accompanying flare-ups differed between groups, with autistic participants more commonly endorsing heightened emotional reactions to sounds, decreased tolerance to non-auditory stimuli, fatigue, and cognitive symptoms, whereas non-autistic adults more frequently reported aural fullness/pressure, continuous ear pain, worsening tinnitus, and middle ear myoclonus. Compared to non-autistic adults, autistic adults also reported a higher incidence of hyperacusis symptoms being exacerbated by stress/anxiety (AUT: 90.5%, NON: 62.5%; OR=5.43 [2.51,12.50]), poor sleep/tiredness (AUT: 82.1%; NON: 60.0%; OR=2.99 [1.50,5.80]), and discomfort from other sensory modalities (AUT: 77.9%, NON: 33.8%; OR=6.70 [3.43, 12.81]). Conclusion: Overall, autistic and non-autistic adults with hyperacusis reported symptoms that were similar in character and severity, and the majority of individuals in both groups endorsed both noise-induced pain and symptom flare-ups. However, autistic adults reported that anxiety and non-auditory stressors play a more substantial role in the generation and exacerbation of symptoms, potentially suggesting a greater non-sensory component in this group. Additional work is needed to better understand the degree to which improving anxiety and/or emotional distress can reduce symptoms of hyperacusis in autistic individuals.
People with misophonia have strong aversive reactions to specific “trigger” sounds. Here we challenge this key idea of specificity. Machine learning was used to identify a misophonic profile from a multivariate sound-response pattern. Misophonia could be classified from most sounds (traditional triggers and non-triggers) and, moreover, cross-classification showed that the profile was largely transferable across sounds (rather than idiosyncratic for each sound). By splitting our participants in other ways, we were able to show—using the same approach—a differential diagnostic profile factoring in potential co-morbidities (autism, hyperacusis, ASMR). The broad autism phenotype was classified via aversions to repetitive sounds rather than the eating sounds most easily classified in misophonia. Within misophonia, the presence of hyperacusis and sound-induced pain had widespread effects across all sounds. Overall, we show that misophonia is characterized by a distinctive reaction to most sounds that ultimately becomes most noticeable for a sub-set of those sounds.
Full-text available
Hyperacusis (decreased sound tolerance) is a prevalent complaint. Yet, to date, no research has qualitatively evaluated the types of problems experienced by adults with hyperacusis. Our service evaluation aims to determine the hyperacusis-related problem domains reported by patients and the degree to which these domains were reported together. Retrospective analysis was conducted on an anonymised clinical dataset from 306 patients who attended a UK tinnitus and hyperacusis treatment centre between 1994 and 2017. Conventional content analysis was used to categorise responses to the question ‘Why is hyperacusis a problem?’ into domains which were then subjected to a cluster analysis. Twenty-five problem domains were identified, of which 12 were further classified into three overarching categories. ‘Fear’, ‘Reduced quality of life’ and ‘Physical reaction to sound’ were most frequently reported problems. Cluster analysis revealed that ‘Sleep difficulties’ and ‘Despondency’, were commonly reported together. Adults with hyperacusis face many challenges in their everyday lives. The nature of these problems indicates the need to develop complex interventions and assessments to aid management of hyperacusis. Current hyperacusis questionnaires may be useful in identifying some problem domains, but further assessment thorough patient interviews is required to fully explore all potential problems and make informed decisions about treatment.
Full-text available
Misophonia is a newly described disorder of sound tolerance characterized by strong negative emotional reactions to specific “trigger” sounds, resulting in significant distress, pathological avoidance, and impairment in daily life. Research on misophonia is still in its infancy, and most existing psychometric tools for assessing misophonia symptoms have not been extensively validated. The purpose of the current study was to introduce and psychometrically validate the duke-vanderbilt Misophonia Screening Questionnaire (DVMSQ), a novel self-report measure of misophonia symptoms that can be used to determine misophonia “caseness” in clinical and research settings. Employing large online samples of general population adults (n = 1403) and adults on the autism spectrum (n = 936), we rigorously evaluated the internal structure, reliability, validity, and measurement invariance of the DVMSQ. Results indicated that 17 of the 20 original DVMSQ items fit well to a bifactor structure with one “general misophonia” factor and four specific factors (anger/aggression, distress/avoidance, impairment, and global impact). DVMSQ total and subscale scores were highly reliable in both general population and autistic adult samples, and the measure was found to be approximately invariant across age, sex, education level, and autism status. DVMSQ total scores also correlated strongly with another measure of misophonia symptoms (Duke Misophonia Questionnaire–Symptom Scale), with correlations between these two measures being significantly stronger than correlations between the DVMSQ and scales measuring other types of sound intolerance (Inventory of Hyperacusis Symptoms [General Loudness subscale] and DSM-5 Severity Measure for Specific Phobia [modified for phonophobia]). Additionally, DVMSQ items were used to operationalize diagnostic criteria for misophonia derived from the Revised Amsterdam Criteria, which were further updated to reflect a recent consensus definition of misophonia (published after the development of the DVMSQ). Using the new DVMSQ algorithm, 7.3% of general population adults and 35.5% of autistic adults met criteria for clinically significant misophonia. Although additional work is needed to further investigate the psychometric properties of the DVMSQ and validate its theory-based screening algorithm using best-estimate clinical diagnoses, this novel measure represents a potentially useful tool to screen for misophonia and quantify symptom severity and impairment in both autistic adults and the general population.
Full-text available
Full-text available
Objective: Hyperacusis, defined as decreased tolerance to sound at levels that would not trouble most individuals, is frequently observed in individuals with autism spectrum disorder (ASD). Despite the functional impairment attributable to hyperacusis, little is known about its prevalence or natural history in the ASD population. The objective of this study was to conduct a systematic review and meta-analysis estimating the current and lifetime prevalence of hyperacusis in children, adolescents, and adults with ASD. By precisely estimating the burden of hyperacusis in the ASD population, the current study aims to enhance recognition of this particular symptom of ASD and highlight the need for additional research into the causes, prevention, and treatment of hyperacusis in persons on the spectrum. Design: We searched PubMed and ProQuest to identify peer-reviewed articles published in English after January 1993. We additionally performed targeted searches of Google Scholar and the gray literature, including studies published through May 2020. Eligible studies included at least 20 individuals with diagnosed ASD of any age and reported data from which the proportion of ASD individuals with current and/or lifetime hyperacusis could be derived. In order to account for multiple prevalence estimates derived from the same samples, we utilized three-level Bayesian random-effects meta-analyses to estimate the current and lifetime prevalence of hyperacusis. Bayesian meta-regression was used to assess potential moderators of current hyperacusis prevalence. In order to reduce heterogeneity due to varying definitions of hyperacusis, we performed a sensitivity analysis on the subset of studies that ascertained hyperacusis status using the Autism Diagnostic Interview-Revised, a structured parent interview. Results: A total of 7783 nonduplicate articles were screened, of which 67 were included in the review and synthesis. Hyperacusis status was ascertained in multiple ways across studies, with 60 articles employing interviews or questionnaires and 7 using behavioral observations or objective measures. The mean (range) age of samples in the included studies was 7.88 (1.00-34.89) years. The meta-analysis of interview/questionnaire measures (k(3) = 103, nASD=13093) estimated the current and lifetime prevalence of hyperacusis in ASD to be 41.42% (95% CrI [37.23, 45.84]) and 60.58% [50.37, 69.76], respectively. A sensitivity analysis restricted to prevalence estimates derived from the ADI-R (k(3) = 25, nASD = 5028) produced similar values. The estimate of current hyperacusis prevalence using objective/observational measures (k(3)= 8, nASD = 488) was 27.30% [14.92, 46.31]. Heterogeneity in the full sample of interview/questionnaire measures was substantial but not significantly explained by any tested moderator. However, prevalence increased sharply with increasing age in studies using the ADI-R (BF10 = 93.10, R2Het = 0.692). Conclusions and Relevance: In this meta-analysis, we found a high prevalence of current and lifetime hyperacusis in individuals with ASD, with a majority of individuals on the autism spectrum experiencing hyperacusis at some point in their lives. The high prevalence of hyperacusis in individuals with ASD across the lifespan highlights the need for further research on sound tolerance in this population and the development of services and/or interventions to reduce the burden of this common symptom.
Full-text available
Atypical behavioral responses to environmental sounds are common in autistic children and adults, with 50-70% of this population exhibiting decreased sound tolerance (DST) at some point in their lives. This symptom is a source of significant distress and impairment across the lifespan, contributing to anxiety, challenging behaviors, reduced community participation, and school/workplace difficulties. However, relatively little is known about its phenomenology or neurocognitive underpinnings. The present article synthesizes a large body of literature on the phenomenology and pathophysiology of DST-related conditions to generate a comprehensive theoretical account of DST in autism. Notably, we argue against conceptualizing DST as a unified construct, suggesting that it be separated into three phenomenologically distinct conditions: hyperacusis (the perception of everyday sounds as excessively loud or painful), misophonia (an acquired aversive reaction to specific sounds), and phonophobia (a specific phobia of sound), each responsible for a portion of observed DST behaviors. We further elaborate our framework by proposing preliminary neurocognitive models of hyperacusis, misophonia, and phonophobia that incorporate neurophysiologic findings from studies of autism.
Full-text available
Many individuals with tinnitus report experiencing hyperacusis (enhanced sensitivity to sounds). However, estimates of the association between hyperacusis and tinnitus is lacking. Here, we investigate this relationship in a Swedish study. A total of 3645 participants (1984 with tinnitus and 1661 without tinnitus) were enrolled via LifeGene, a study from the general Swedish population, aged 18-90 years, and provided information on socio-demographic characteristics, as well as presence of hyperacusis and its severity. Tinnitus presence and severity were self-reported or assessed using the Tinnitus Handicap Inventory (THI). Phenotypes of tinnitus with (n = 1388) or without (n = 1044) hyperacusis were also compared. Of 1661 participants without tinnitus, 1098 (66.1%) were women and 563 were men (33.9%), and the mean (SD) age was 45.1 (12.9). Of 1984 participants with tinnitus, 1034 (52.1%) were women and 950 (47.9%) were men, and the mean (SD) age was 47.7 (14.0) years. Hyperacusis was associated with any tinnitus [Odds ratio (OR) 3.51, 95% confidence interval (CI) 2.99-4.13], self-reported severe tinnitus (OR 7.43, 95% CI 5.06-10.9), and THI ≥ 58 (OR 12.1, 95% CI 7.06-20.6). The association with THI ≥ 58 was greater with increasing severity of hyperacusis, the ORs being 8.15 (95% CI 4.68-14.2) for moderate and 77.4 (95% CI 35.0-171.3) for severe hyperacusis. No difference between sexes was observed in the association between hyperacusis and tinnitus. The occurrence of hyperacusis in severe tinnitus is as high as 80%, showing a very tight relationship. Discriminating the pathophysiological mechanisms between the two conditions in cases of severe tinnitus will be challenging, and optimized study designs are necessary to better understand the mechanisms behind the strong relationship between hyperacusis and tinnitus.
Full-text available
Objective Analyze a large sample with detailed clinical data of misophonia subjects in order to determine the psychiatric, somatic and psychological nature of the condition. Methods This observational study of 779 subjects with suspected misophonia was conducted from January 2013 to May 2017 at the outpatient-clinic of the Amsterdam University Medical Centers, location AMC, the Netherlands. We examined DSM-IV diagnoses, results of somatic examination (general screening and hearing tests), and 17 psychological questionnaires (e.g., SCL-90-R, WHOQoL). Results The diagnosis of misophonia was confirmed in 575 of 779 referred subjects (74%). In the sample of misophonia subjects (mean age, 34.17 [SD = 12.22] years; 399 women [69%]), 148 (26%) subjects had comorbid traits of obsessive-compulsive personality disorder, 58 (10%) mood disorders, 31 (5%) attention-deficit (hyperactivity) disorder, and 14 (3%) autism spectrum conditions. Two percent reported tinnitus and 1% hyperacusis. In a random subgroup of 109 subjects we performed audiometry, and found unilateral hearing loss in 3 of them (3%). Clinical neurological examination and additional blood test showed no abnormalities. Psychological tests revealed perfectionism (97% CPQ>25) and neuroticism (stanine 7 NEO-PI-R). Quality of life was heavily impaired and associated with misophonia severity (rs (184) = -.34 p = < .001, p = < .001). Limitations This was a single site study, leading to possible selection–and confirmation bias, since AMC-criteria were used. Conclusions This study with 575 subjects is the largest misophonia sample ever described. Based on these results we propose a set of revised criteria useful to diagnose misophonia as a psychiatric disorder.
Full-text available
Current computations of commonly used fit indices in structural equation modeling (SEM), such as RMSEA and CFI, indicate much better fit when the data are categorical than if the same data had not been categorized. As a result, researchers may be led to accept poorly fitting models with greater frequency when data are categorical. In this article, I first explain why the current computations of categorical fit indices lead to this problematic behavior. I then propose and evaluate alternative ways to compute fit indices with categorical data. The proposed computations approximate what the fit index values would have been had the data not been categorized. The developments in this article are for the DWLS (diagonally weighted least squares) estimator, a popular limited information categorical estimation method. I report on the results of a simulation comparing existing and newly proposed categorical fit indices. The results confirmed the theoretical expectation that the new indices better match the corresponding values with continuous data. The new fit indices performed well across all studied conditions, with the exception of binary data at the smallest studied sample size (N = 200), when all categorical fit indices performed poorly.
Full-text available
Turmoil has engulfed psychological science. Causes and consequences of the reproducibility crisis are in dispute. With the hope of addressing some of its aspects, Bayesian methods are gaining increasing attention in psychological science. Some of their advantages, as opposed to the frequentist framework, are the ability to describe parameters in probabilistic terms and explicitly incorporate prior knowledge about them into the model. These issues are crucial in particular regarding the current debate about statistical significance. Bayesian methods are not necessarily the only remedy against incorrect interpretations or wrong conclusions, but there is an increasing agreement that they are one of the keys to avoid such fallacies. Nevertheless, its flexible nature is its power and weakness, for there is no agreement about what indices of “significance” should be computed or reported. This lack of a consensual index or guidelines, such as the frequentist p-value, further contributes to the unnecessary opacity that many non-familiar readers perceive in Bayesian statistics. Thus, this study describes and compares several Bayesian indices, provide intuitive visual representation of their “behavior” in relationship with common sources of variance such as sample size, magnitude of effects and also frequentist significance. The results contribute to the development of an intuitive understanding of the values that researchers report, allowing to draw sensible recommendations for Bayesian statistics description, critical for the standardization of scientific reporting.
Full-text available
Objective To determine research priorities in hyperacusis that key stakeholders agree are the most important. Design/setting A priority setting partnership using two international surveys, and a UK prioritisation workshop, adhering to the six-staged methodology outlined by the James Lind Alliance. Participants People with lived experience of hyperacusis, parents/carers, family and friends, educational professionals and healthcare professionals who support and/or treat adults and children who experience hyperacusis, including but not limited to surgeons, audiologists, psychologists and hearing therapists. Methods The priority setting partnership was conducted from August 2017 to July 2018. An international identification survey asked respondents to submit any questions/uncertainties about hyperacusis. Uncertainties were categorised, refined and rephrased into representative indicative questions using thematic analysis techniques. These questions were verified as ‘unanswered’ through searches of current evidence. A second international survey asked respondents to vote for their top 10 priority questions. A shortlist of questions that represented votes from all stakeholder groups was prioritised into a top 10 at the final prioritisation workshop (UK). Results In the identification survey, 312 respondents submitted 2730 uncertainties. Of those uncertainties, 593 were removed as out of scope, and the remaining were refined into 85 indicative questions. None of the indicative questions had already been answered in research. The second survey collected votes from 327 respondents, which resulted in a shortlist of 28 representative questions for the final workshop. Consensus was reached on the top 10 priorities for future research, including identifying causes and underlying mechanisms, effective management and training for healthcare professionals. Conclusions These priorities were identified and shaped by people with lived experience, parents/carers and healthcare professionals, and as such are an essential resource for directing future research in hyperacusis. Researchers and funders should focus on addressing these priorities.
Background Hyperacusis is a prevalent auditory disorder that causes significant distress and negatively affects quality of life for many patients. Patients with hyperacusis often have different complaints about the sounds and situations that they experience. Audiologists may have few patients with hyperacusis, and a limited understanding of the sounds and situations that are reported to be challenging by their patients. Purpose To investigate the common complaints reported by hyperacusis patients. Research design A qualitative study was conducted with 11 hyperacusis patients who participated in a group session. Results All 11 hyperacusis patients experienced negative reactions to specific sounds. In addition, many patients reported physical symptoms such as headaches, balance problems, dysosmia (strong smell problems), and light sensitivity. Sounds that induced discomfort were wide ranging and included low-frequency sounds, high-frequency sounds, wide-band noise, and sudden, high-intensity sounds. Most patients (9/11, 81.8%) reported negative reactions to music in loud rock concerts. Patients reported that stress/tension (90.9%) worsened their hyperacusis, while removing themselves from noise (90.9%) relieved their hyperacusis. Conclusion Loudness is only one of the many factors related to the discomfort of patients with hyperacusis. Across patients, we observed that there were different complaints about the sounds and situations that produced difficulty due to hyperacusis. Physical symptoms following sound exposure were also reported by the patients, suggesting that hyperacusis is a complex disorder and requires intervention that often involves multiple members of the medical team.
Despite an ongoing stream of lamentations, many empirical disciplines still treat the p value as the sole arbiter to separate the scientific wheat from the chaff. The continued reign of the p value is arguably due in part to a perceived lack of workable alternatives. In order to be workable, any alternative methodology must be (1) relevant: it has to address the practitioners’ research question, which—for better or for worse—most often concerns the test of a hypothesis, and less often concerns the estimation of a parameter; (2) available: it must have a concrete implementation for practitioners’ statistical workhorses such as the t test, regression, and ANOVA; and (3) easy to use: methods that demand practitioners switch to the theoreticians’ programming tools will face an uphill struggle for adoption. The above desiderata are fulfilled by Harold Jeffreys’s Bayes factor methodology as implemented in the open-source software JASP. We explain Jeffreys’s methodology and showcase its practical relevance with two examples.
Objectives To evaluate the efficacy of a multi-modal migraine prophylaxis therapy for patients with hyperacusis. Methods In a prospective cohort, patients with hyperacusis were treated with a multi-modal step-wise migraine prophylactic regimen (nortriptyline, verapamil, topiramate, or a combination thereof) as well as lifestyle and dietary modifications. Pre- and post-treatment average loudness discomfort level (LDL), hyperacusis discomfort level measured by a visual analogue scale (VAS), and scores on the modified Khalfa questionnaire for severity of hyperacusis were compared. Results Twenty-two of the 25 patients (88%) reported subjective resolution of their symptoms following treatment. Post-treatment audiograms showed significant improvement in average LDL from 81.3 ± 3.2 dB to 86.4 ± 2.6 dB ( P < .001), indicating increased sound tolerability. The VAS discomfort level also showed significant improvement from a pre-treatment average of 7.7 ± 1.1 to 3.7 ± 1.6 post-treatment ( P < .001). There was also significant improvement in the average total score on modified Khalfa questionnaire (32.2 ± 3.6 vs 22.0 ± 5.7, P < .001). Conclusions The majority of patients with hyperacusis demonstrated symptomatic improvement from migraine prophylaxis therapy, as indicated by self-reported and audiometric measures. Our findings indicate that, for some patients, hyperacusis may share a pathophysiologic basis with migraine disorder and may be successfully managed with multimodal migraine prophylaxis therapy.