Hearing Loss, Age-Related
Conditions
Keywords
Hearing loss, Speech processing, Speech-in-noise, Aging
Brief summary
Mild to moderate hearing loss remains undertreated, largely because of the high cost of hearing aids. A promising and much less expensive alternative is the use of personal sound amplification products (PSAPs), which are electronic, portable, over-the-counter devices that amplify sound. Studies have shown that the use of PSAPs provides significant hearing benefits and improves the quality of life for older adults with mild to moderate hearing loss. However, there is insufficient data to determine the impact of PSAPs use on speech processing in the brain. The purpose of this study is to use electroencephalography (EEG) measurements to assess the neurobiological and behavioral effects of PSAPs on speech perception in noise in individuals with mild to moderate hearing loss. The investigators expect that the PSAPs use will result in an immediate improvement in the ability to perceive speech-in-noise, supporting that these hearing devices may be a means of restoring communication skills in people with mild to moderate hearing loss. Behavioral benefits will be associated with increased brain activity in auditory regions and connectivity between auditory and speech regions in the brain.
Detailed description
This study will consist of two sessions of 3 hours each. On one session, participants will perform the speech-in-noise task without hearing devices and on the other session, participants will perform the speech-in-noise task while wearing personal sound amplification products. The order of the sessions will be counterbalanced across participants. The speech-in-noise task consists in a word discrimination task in babble noise at three signal-to-noise ratios. On each trial, the task is to determine whether two words are identical or different.
Interventions
DEVICEPersonal sound amplification products
Participants will be tested with bilateral personal sound amplification products.
Sponsors
Rotman Research Institute at Baycrest
Study design
Allocation
RANDOMIZED
Intervention model
CROSSOVER
Primary purpose
BASIC_SCIENCE
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
60 Years to 90 Years
Healthy volunteers
No
Inclusion criteria
* Right-handed
Exclusion criteria
* Mother tongue not English * Language impairment * Dementia * Cerebrovascular diseases * Untreated vision impairment; * Tinnitus and otologic disorders * Cochlear implant * History of prior hearing aid use * Diagnosed addiction (alcohol or drugs) * Significant medical or neurocognitive conditions or interventions likely to significantly impact cognitive function (e.g., epilepsy, stroke, traumatic brain injury with loss of consciousness \> 5 minutes, brain tumor, multiple sclerosis, hepatitis C, developmental delay, electroconvulsive therapy) * a diagnosis (based on the Diagnostic and Statistical Manual of Mental Disorders (DSM-5)) of major depressive disorder with active symptoms within 90 days of study entry, past or present psychosis, or other psychiatric disorders such as obsessive-compulsive disorder, generalized anxiety disorder, and bipolar disorder
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Quick Speech In Noise Score With and Without PSAPs | 1 hour | Performance on the QuickSIN when using PSAPs and when using no device. The QuickSIN measures the signal-to-noise ratio loss. A high score indicates poorer speech understanding in noise. |
| Percentage of Correct Responses With and Without PSAPs | 1 hour | Percentage (%) of correct responses in the speech-in-noise task when using PSAPs and when using no device. |
| Reaction Time With and Without PSAPs | 1 hour | Reaction time (in milliseconds) in the speech-in-noise task when using PSAPs and when using no device. |
| Alpha Power (8-12 Hz) During the Processing of the First Word in a High Noise Condition, With and Without PSAPs | 1 hour | Alpha power (8-12 Hz) for a cluster of temporoparietal electrodes (T7, TP7, CP5), measured between 50 and 500 ms after the onset of the first word of the speech-in-noise task, under a high noise condition (signal-to-noise ratio of -3 dB), with and without PSAPs. Alpha power was measured using temporal spectral evolution analysis with EEG. Values are baseline corrected with the pre-stimulus interval period (-500, 0 ms). Higher values indicate greater alpha power compared to the pre-stimulus interval period (i.e., event-related synchronization). |
| Alpha Power (8-12 Hz) During the Processing of the First Word in a Low Noise Condition, With and Without PSAPs | 1 hour | Alpha power (8-12 Hz) for a cluster of temporoparietal electrodes (T7, TP7, CP5), measured between 50 and 500 ms after the onset of the first word of the speech-in-noise task, under a low noise condition (signal-to-noise ratio of +3 dB), with and without PSAPs. Alpha power was measured using temporal spectral evolution analysis with EEG. Values are baseline corrected with the pre-stimulus interval period (-500, 0 ms). Higher values indicate greater alpha power compared to the pre-stimulus interval period (i.e., event-related synchronization). |
| Alpha Power (8-12 Hz) During the Processing of the First Word in a Medium Noise Condition, With and Without PSAPs | 1 hour | Alpha power (8-12 Hz) for a cluster of temporoparietal electrodes (T7, TP7, CP5), measured between 50 and 500 ms after the onset of the first word of the speech-in-noise task, under a medium noise condition (signal-to-noise ratio of 0 dB), with and without PSAPs. Alpha power was measured using temporal spectral evolution analysis with EEG. Values are baseline corrected with the pre-stimulus interval period (-500, 0 ms). Higher values indicate greater alpha power compared to the pre-stimulus interval period (i.e., event-related synchronization). |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Self-reported Measure of Listening Effort With and Without PSAPs | 1 hour | Self-reported listening effort in the speech-in-noise task when using PSAPs and when using no device. Participants used a seven-point Likert scale to rate the listening effort required to complete the speech-in-noise task. The specific question was: Using the scale in front of you, can you estimate how much effort it took you to understand the words in the presence of background noise? If you think that the amount of effort was between two numbers on the scale, it is fine for you to pick a fraction, with number 1 corresponding to No effort and number 7 corresponding to Extreme effort. A high score indicates a greater listening effort. |
Countries
Canada
Participant flow
Recruitment details
72 participants were screened for eligibility between March 1, 2022 and September 31, 2022 by telephone interview at Baycrest Hospital in Toronto, Canada.
Pre-assignment details
32 of the 72 participants were randomized. Of those who were not randomized, 22 did not meet the inclusion criteria, 14 declined to participate and four had other reasons.
Participants by arm
| Arm | Count |
|---|---|
| Personal Sound Amplification Products First, Then Control Participants were tested for speech-in-noise perception while using bilateral personal sound amplification products during the first session. After a one-week washout period, they were retested for speech-in-noise perception while not using a device. | 15 |
| Control First, Then Personal Sound Amplification Products Participants were tested for speech perception in noise while using no devices during the first session. After a one-week washout period, they were retested for speech-in-noise perception while using bilateral personal sound amplification products. | 13 |
| Total | 28 |
Withdrawals & dropouts
| Period | Reason | FG000 | FG001 |
|---|---|---|---|
| Washout (1 Week) | Withdrawal by Subject | 1 | 3 |
Baseline characteristics
| Characteristic | Personal Sound Amplification Products First, Then Control | Control First, Then Personal Sound Amplification Products | Total |
|---|---|---|---|
| Age, Continuous | 74 years STANDARD_DEVIATION 6.6 | 72.23 years STANDARD_DEVIATION 6.25 | 73.18 years STANDARD_DEVIATION 6.38 |
| Race and Ethnicity Not Collected | — | — | 0 Participants |
| Region of Enrollment Canada | 15 participants | 13 participants | 28 participants |
| Sex: Female, Male Female | 9 Participants | 10 Participants | 19 Participants |
| Sex: Female, Male Male | 6 Participants | 3 Participants | 9 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | 0 / 28 | 0 / 28 |
| other Total, other adverse events | 0 / 28 | 0 / 28 |
| serious Total, serious adverse events | 0 / 28 | 0 / 28 |
Outcome results
Primary
Alpha Power (8-12 Hz) During the Processing of the First Word in a High Noise Condition, With and Without PSAPs
Alpha power (8-12 Hz) for a cluster of temporoparietal electrodes (T7, TP7, CP5), measured between 50 and 500 ms after the onset of the first word of the speech-in-noise task, under a high noise condition (signal-to-noise ratio of -3 dB), with and without PSAPs. Alpha power was measured using temporal spectral evolution analysis with EEG. Values are baseline corrected with the pre-stimulus interval period (-500, 0 ms). Higher values indicate greater alpha power compared to the pre-stimulus interval period (i.e., event-related synchronization).
Time frame: 1 hour
Population: One participant did not complete the EEG portion of the study. Two other participants were excluded due to noisy data.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Personal Sound Amplification Products | Alpha Power (8-12 Hz) During the Processing of the First Word in a High Noise Condition, With and Without PSAPs | -0.08 percentage of power change | Standard Deviation 0.11 |
| Control | Alpha Power (8-12 Hz) During the Processing of the First Word in a High Noise Condition, With and Without PSAPs | -0.01 percentage of power change | Standard Deviation 0.13 |
Comparison: Null hypothesis is that there was no difference in alpha power between PSAPs and Control. Data were analyzed using a linear mixed-effects (LME) model. The model included session (PSAPs, Control) and signal-to-noise ratio (SNR) as within-subject factors. Participants were included as a random factor.p-value: 0.36Mixed Models Analysis
Primary
Alpha Power (8-12 Hz) During the Processing of the First Word in a Low Noise Condition, With and Without PSAPs
Alpha power (8-12 Hz) for a cluster of temporoparietal electrodes (T7, TP7, CP5), measured between 50 and 500 ms after the onset of the first word of the speech-in-noise task, under a low noise condition (signal-to-noise ratio of +3 dB), with and without PSAPs. Alpha power was measured using temporal spectral evolution analysis with EEG. Values are baseline corrected with the pre-stimulus interval period (-500, 0 ms). Higher values indicate greater alpha power compared to the pre-stimulus interval period (i.e., event-related synchronization).
Time frame: 1 hour
Population: One participant did not complete the EEG portion of the study. Two other participants were excluded due to noisy data.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Personal Sound Amplification Products | Alpha Power (8-12 Hz) During the Processing of the First Word in a Low Noise Condition, With and Without PSAPs | -0.10 percentage of power change | Standard Deviation 0.11 |
| Control | Alpha Power (8-12 Hz) During the Processing of the First Word in a Low Noise Condition, With and Without PSAPs | -0.12 percentage of power change | Standard Deviation 0.12 |
Comparison: Null hypothesis is that there was no difference in alpha power between PSAPs and Control. Data were analyzed using a linear mixed-effects (LME) model. The model included session (PSAPs, Control) and signal-to-noise ratio (SNR) as within-subject factors. Participants were included as a random factor.p-value: 1Mixed Models Analysis
Primary
Alpha Power (8-12 Hz) During the Processing of the First Word in a Medium Noise Condition, With and Without PSAPs
Alpha power (8-12 Hz) for a cluster of temporoparietal electrodes (T7, TP7, CP5), measured between 50 and 500 ms after the onset of the first word of the speech-in-noise task, under a medium noise condition (signal-to-noise ratio of 0 dB), with and without PSAPs. Alpha power was measured using temporal spectral evolution analysis with EEG. Values are baseline corrected with the pre-stimulus interval period (-500, 0 ms). Higher values indicate greater alpha power compared to the pre-stimulus interval period (i.e., event-related synchronization).
Time frame: 1 hour
Population: One participant did not complete the EEG portion of the study. Two other participants were excluded due to noisy data.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Personal Sound Amplification Products | Alpha Power (8-12 Hz) During the Processing of the First Word in a Medium Noise Condition, With and Without PSAPs | -0.09 percentage of power change | Standard Deviation 0.14 |
| Control | Alpha Power (8-12 Hz) During the Processing of the First Word in a Medium Noise Condition, With and Without PSAPs | -0.08 percentage of power change | Standard Deviation 0.15 |
Comparison: Null hypothesis is that there was no difference in alpha power between PSAPs and Control. Data were analyzed using a linear mixed-effects (LME) model. The model included session (PSAPs, Control) and signal-to-noise ratio (SNR) as within-subject factors. Participants were included as a random factor.p-value: 1Mixed Models Analysis
Primary
Percentage of Correct Responses With and Without PSAPs
Percentage (%) of correct responses in the speech-in-noise task when using PSAPs and when using no device.
Time frame: 1 hour
Population: All participants who completed all study visits were included in the analysis.
| Arm | Measure | Value (MEAN) |
|---|---|---|
| Personal Sound Amplification Products | Percentage of Correct Responses With and Without PSAPs | 85.2 percentage of correct answers |
| Control | Percentage of Correct Responses With and Without PSAPs | 84.0 percentage of correct answers |
Comparison: Null hypothesis is that there was no difference in percentage of correct answers between PSAPs and Control. Data were analyzed using a linear mixed-effects (LME) model. The model included session (PSAPs, Control) and signal-to-noise ratio (SNR) as within-subject factors. Participants were included as a random factor.p-value: 0.03Mixed Models Analysis
Primary
Quick Speech In Noise Score With and Without PSAPs
Performance on the QuickSIN when using PSAPs and when using no device. The QuickSIN measures the signal-to-noise ratio loss. A high score indicates poorer speech understanding in noise.
Time frame: 1 hour
Population: All participants who completed all study visits were included in the analysis.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Personal Sound Amplification Products | Quick Speech In Noise Score With and Without PSAPs | 3.6 dB | Standard Deviation 2.2 |
| Control | Quick Speech In Noise Score With and Without PSAPs | 4.6 dB | Standard Deviation 2.4 |
Comparison: Null hypothesis is that there was no difference in Quick Speech In Noise score between PSAPs and Control. The performance of the QuickSIN for the two sessions (PSAPs, Control) was compared using the Wilcoxon signed rank test on paired samples.p-value: 0.005Wilcoxon (Mann-Whitney)
Primary
Reaction Time With and Without PSAPs
Reaction time (in milliseconds) in the speech-in-noise task when using PSAPs and when using no device.
Time frame: 1 hour
Population: All participants who completed all study visits were included in the analysis. Data from one participant was lost due to a technical issue. Additionally, one participant was excluded because their reaction time values exceeded 3 times the interquartile range.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Personal Sound Amplification Products | Reaction Time With and Without PSAPs | 528.7 milliseconds | Standard Deviation 155.1 |
| Control | Reaction Time With and Without PSAPs | 502.0 milliseconds | Standard Deviation 145.9 |
Comparison: Null hypothesis is that there was no difference in reaction time between PSAPs and Control. Data were analyzed using a linear mixed-effects (LME) model. The model included session (PSAPs, Control) and signal-to-noise ratio as within-subject factors. Participants were included as a random factor.p-value: 0.28Mixed Models Analysis
Secondary
Self-reported Measure of Listening Effort With and Without PSAPs
Self-reported listening effort in the speech-in-noise task when using PSAPs and when using no device. Participants used a seven-point Likert scale to rate the listening effort required to complete the speech-in-noise task. The specific question was: Using the scale in front of you, can you estimate how much effort it took you to understand the words in the presence of background noise? If you think that the amount of effort was between two numbers on the scale, it is fine for you to pick a fraction, with number 1 corresponding to No effort and number 7 corresponding to Extreme effort. A high score indicates a greater listening effort.
Time frame: 1 hour
Population: All participants who completed all study visits were included in the analysis.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Personal Sound Amplification Products | Self-reported Measure of Listening Effort With and Without PSAPs | 3.8 score on a scale | Standard Deviation 1.1 |
| Control | Self-reported Measure of Listening Effort With and Without PSAPs | 4.2 score on a scale | Standard Deviation 0.9 |
Comparison: Null hypothesis is that there was no difference in listening effort between PSAPs and Control. The listening effort score was analyzed using a cumulative link mixed model. A logit link function with flexible thresholds was used. The model included Session (PSAPs, Control) and Block (1, 2, 3) as within-subject factors. Participants were included as a random factor.p-value: <0.001Cumulative link mixed model