AUDI 556 - Midterm

3 forms of assessment

auditory assessment - test battery

auditory needs assessment

non-auditory assessment

7 considerations for candidacy

1. type of loss

2. amount of loss

3. slope of loss

4. word recognition

5. dynamic range

6. non-auditory characteristics

7. red flags

How does type of loss affect candidacy?

CHL - requires medical clearance to determine if loss is permanent

SNHL - amplification does not always solve problems of discrimination

How does amount of loss affect candidacy?

Any degree of loss can be a candidate for amplification

• consider their PTA and speech range (500-3000 Hz)

Borderline/mild loss → consider motivation and listening needs

Moderate → tend to be good candidates

Severe/profound → limited benefit of amplification supplemented with counselling and other comm. strategies

How does slope of HL affect candidacy?

Presents a balancing act for fitting hearing aids - consider other factors to decide if amplification is necessary / beneficial

• it is difficult to only amplify high frequencies

• feedback becomes an issue

Issues with using WRS for candidacy

Large test-retest variability

Signal is not frequency-shaped - cannot match to their audiogram

No cues given - does not reflect real world

Very little evidence between WRS and self-reported HA satisfaction

How does dynamic range affect candidacy?

individuals with HL have a reduced dynamic range

• squishing the signal into this range can create distortion

• fitting for profound loss may create lots of distortion

What other factors are considered in candidacy?

Motivation

Communication needs

Expectations

Cognitive ability - can they care for HA and use them

Cost - not our decision to make

Dexterity & vision

Ear canal (health, shape, size)

RED FLAGS FOR CANDIDACY (11)

1. Sudden/progressive/fluctuating HL

2. Draining/bleeding within last 90 days

3. Pain or discomfort in ear

4. Unilateral/pulsatile tinnitus

5. Acute, recurring or chronic dizziness

6. Foreign object in ear canal

7. Trauma or unexplained abnormality

8. Asymmetry greater than 30 dB at more than 1 freq.

9. ABG >= 15 dB at 0.5, 1 AND 2 kHz

10. 40% difference in WRS with symmetrical hearing on a 25 word recorded list

11. Facial nerve paralysis

Steps of Verification & Selection

1. Select characteristics of amplification

   • physical & electroacoustic

2. Quality control (ANSI)

3. Fitting & verification

4. Hearing assistive technology

Elements of Instruction & Counselling

Care and use

Realistic expectations - what can HA do / not do

Counselling

Aural rehabilitation

Intake interview gives us

Client’s perception of their HL

Expectations, motivation for amplification, lifestyle, degree of disability

Heads-up for need for medical referral

Qualifications for third-party funding

Perspective of companion (if they bring someone)

Opportunity to build rapport

Goals for use in validation of benefit (COSI)

Speech in Noise Measures

Objective: quickSIN, HINT, WIN, SIN, SPIN, CST

Subjective: Acceptable noise level

SNR Loss

can predict some of the variance in success with HA

can help choose technology (not researched)

• 0-2 dB SNR loss - normal

• 3-6 dB - use directional mics

• 7-12 dB - consider remote mic

  • 12+ dB - remote mic is necessity

Acceptable Noise Level (ANL)

noise intensity that’s acceptable to the listener while trying to understand speech

• play recording of speech at MCL, add background noise and increase level until it is at max that listener can follow words of story

bigger ANL - less likely to succeed with HA

What does Davidson et al. 2021 show?

ANL and speech recognition in noise are both sometimes related to HA satisfaction

• weak-moderate evidence

• stronger relationship than in quiet

Measures of loudness (4)

LDL - loudness discomfort level

UCL - uncomfortable level

ULC - upper limit of comfort

• about 2 dB below LDL

ULCL - upper limit of comfortable listening

LDL administration variation

LDL measurement procedures can differ on

• instruction

• stimuli

• psychophysical procedure

• stimulus delivery method

Hawkins Pascoe LDL

validated with high reliability

• less than 4 dB shift across 4 days

Why do we need acoustic transforms

HL thresholds are referenced to dB SPL measured in a coupler

• real ears are not similar to a coupler or to each other

• HL thresholds are not standard and cannot tell us the lowest SPL value someone can hear

• HA are measured in SPL - we need to make them match

Auditory needs assessments measure: (4)

Communication needs

Realistic expectations

Goals

Extent of perceived impairment

Communication Needs/Goals assessments (4)

APHAB - abbreviated profile of HA benefit

COSI - client oriented scale of improvement

GHAPB - Glasgow HA profile of benefit

HHIE - hearing handicap inventory for the elderly

Expectation assessments (2)

ECHO - expected consequences of HA ownership

SADL - satisfaction with amplification in daily life (but in future tense)

Extent of Perceived Impairment

self-perceived loss is a good predictor of hearing aid uptake

Non-Auditory Assessment Elements

Cognition

Manual dexterity

Vision

General health

Support system

Work and recreation

Personality

Prior experience

Motivation

COAT

Characteristics of amplification tool

• gather info about values and preferences that will allow the AUD to make recommendations re: style and technology

What does prior experience tell us?

If someone has previously had hearing aids, what was their experience?

How closely should you match their former HA in terms of processing, gain, output

RETSPL

reference equivalent threshold in SPL

RECD

real ear to coupler difference

REDD

real ear to dial difference

(RETSPL + RECD)

How to calculate ear canal thresholds

dB HL + RETSPL + Custom RECD

RECD Process

1. Calibrate / measure in coupler

2. Perform otoscopy and insert probe tube

3. Use insert earphones to play signal

4. Verifit measures the RECD

REUR

real ear unaided response (SPL)

REUG

real ear unaided gain

Factors affecting RECD

Sources of resonance - ear canal and pinna

Probe tube insertion - further away from TM, start to lose high freq.

Calibration

Noise - need noise floor to be low enough to test

Loudspeaker location - follow manual instructions

Substitution Calibration

single mic calibrates for specific spot

• not ideal for REM - we may need to measure both ears

Concurrent Calibration

adjusts sound source throughout measurement

• continual feedback loop

Stored Calibration

equalize once before measuring, correction values are stored and applied ot every measure

• if anything changes in environment, we need to redo calibration

earmold styles (3)

full shell

skeleton

canal

hearing aid styles

BTE, RITE/RIC, RITA, ITE, ITC, CIC IIC

purpose of ANSI testing

ensure that HA are up to specs - do what we want them to do

Physical (style) Selection Considerations

Power - larger = greater power

Battery size - large HA = larger battery

Earmold or receiver options

• occlusion (low freq. needs)

• ear anatomy (size and shape)

Non-auditory issues such as dexterity & vision

Features that vary by style

• directional mics

• t-coils

• compatibility with wireless connectivity

Feature (tech) Selection Considerations

volume control

• does the person need/want control

• automatic = higher tech level

T-coils

• program/space requirements

wireless compatibility and accessories

• phone/streaming needs

• remote mic / FM

directional mics

noise reduction

multiple memories

feedback management

Validated prescriptive methods provide

Calculated gain and output that provide

• comfort of loud sounds

• audibility of average speech

• optimal speech recognition

• free of distortion

• wide bandwidth (freq range)

• wide input range audible and comfortable

sensation level

dB above threshold level needed to comfortably listen to speech

headroom

dB difference between amplified signal and MPO

• ensures we don’t exceed ULC

• analogy: truck going under over-pass requires headroom

dynamic range

dB range between threshold and ULC

NAL Linear Amplification Prescriptions

NAL - original

NAL-R - ½ gain

NAL-RP - 2/3 gain

NAL non-linear amplification prescriptions

NAL-NL1 - compatible with compression

NAL-NL2 - revised version with field data

NAL-NL3 - focus on personalization, targets for hearing in noise

Aims of NAL

maximize speech intelligibility

equalize loudness across frequency bands

DSL history

Linear prescription - one set of gain targets for children

• Does this sound better or does this

Non-linear - input/output gain prescription

• m[i/o] - multistage input output

Goals of DSL

Comfort

Intelligibility of average speech

Audibility of broad frequency range

Audibility of broad input range

Use with all circuit types (linear/compression, bone anchored HA)

Comparing NAL and DSL

No functional difference between prescriptive methods

• APHAB questionnaire had no significant differences

Some minor gain differences at frequency extremes (low and high ends)

Why not use manufacturer’s prescription?

Proprietary methods that are built into software do not have known rationale/goals

Targets are not available in the real-ear measurement system, so we can’t verify them

Ways to get targets

Calculate by hand

Use software programs (standalone software)

Built into real-ear systems

Manufacturer’s NOAH modules

Pre-fitting tasks

Connect HA

Set acoustic parameters (prescription)

• set to 100% gain

• select receiver and vent options

Calibrate test box

Calibrate on ear module

Send audiogram to VeriFit

Calibrate RECD

Fitting tasks

HA orientation/counselling

Put HA in and ask how it sounds

Perform feedback test

Run RECD

Run REM

• average, soft, loud and MPO

Purpose of HA verification

Optimize speech signal (loudness, intelligibility, quality) for an individual with HL based on their goals

Ensure hearing aid is doing what we want it to do for the individual

Why not set and forget?

1. HA vary from specs - need to ensure it is within tolerance and get exact values

2. Real ears vary - lots of variation in size and shape → different resonance

How similar are proprietary algorithms from different manufacturers?

Shane Moodie study

21 dB output range for average speech

26 dB for loud speech

30 dB range for MPO

Should we use pleasantness as a measure for HA?

No because clients tend to rate the same sound level they came in with as more pleasant

• prefer the same rather than new change

Valente’s 7 recommendations for verification

1. Choice of assessment signal

2. Physical fit

3. Occlusion effect

4. Gain verification

5. Output verification

6. Aided sound field threshold

7. Verification of special features

Signal levels in verification

Soft (50-55 dB), average (60-65 dB), loud (70-75 dB) and MPO (85-90 dB)

Signal types in verification

Pure tones - MPO

Speech noise

Speech

Loudspeaker location in verification

Check manual for real ear system to determine best location

• typically in front or at 45 degrees

• within 1-1.5 ft of person

Physical fit considerations in verification

Assess the following when HA is inserted first time - if there is an issue fix before continuing with fitting

Subjective comfort/fit

Appearance

Microphone angle

Ease of insertion/removal

Any leak/feedback

Occlusion effect in verification

Unnatural sound of one’s own voice resulting from occluding of the external ear canal

• subjective assessment - ask how it sounds

• objective assessment - probe mic measures voice in ear canal in closed vs open fitting

Gain verification order

Average speech

Soft speech

Loud speech

Order is based on efficiency - intelligibility of average speech is primary goal

• if we are off target for average input, we are likely off target for all

REIR

Real ear insertion response

REIG

Real ear insertion gain

Gain of HA relative to unoccluded ear (REUR)

REAR

real ear aided response

frequency response of HA measured in ear canal

REAG

real ear aided gain

gain of HA, pinna and head relative to soundfield

How to calculate REAR

REAR - REUG

Output verification methods

1. RESR - real ear saturation response / MPO

2. OSPL 90 + RECD

3. Aided LDLs

Aided soundfield threshold testing

Perform threshold search in sound field with one HA at a time, other ear occluded

Pros:

• aided vs unaided shows that HAs are making a difference

• may be the only option, such as with extended-wear HAs

Cons:

• large confidence interval - can’t be sure changes are due to HA

• only testing at threshold, w/ discrete stimuli

• takes lots of time

What special features can we do real-ear verification on

directional mic

telecoil

noise reduction

feedback reduction

frequency lowering

Open fitting considerations in verification

We have to measure verification in the real ear, not in the coupler

Must turn off reference mic during test

• cannot use concurrent calibration - amplified sound will be picked up by reference mic, which will then adjust the signal

REOR

real ear occluded response

SPL at eardrum with HA in place and turned off when external signal is presented

REOG

real ear occluded gain (more like a loss though)

how much attenuation occurs due to occlusion of ear

• eg. REOG of earplug is ~-15dB

• eg. vented HA has no loss in low frequencies

When to verify targets?

Initial fitting - set to target

Final fitting - after adjusting to client preferences

When client’s thresholds have changed (20dB or more at 2 freq)

Recommended

• when there have been changes in venting

• after making new earmold

• after HA repair

Benefits of performing REM

Increased:

• self-reported listening ability

• speech intelligibility in quiet

• speech intelligibility in noise

• preference

No difference in sound quality

Fewer return visits, higher reports of success when clients recall REM

Audioscan SII

Speech intelligibility index targets provides guide for goodness of fit

• function of threshold

If we are within target range, we can feel comfortable with fitting

• exception: severe/profound HL clients

Audioscan RMSE

Root mean square error helps us tell how far off target we are

• helps us to decide when to stop fiddling

• target - within 5 dB