Variability In Hearing Aid Outcomes In Older Adults: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Variability In Hearing Aid Outcomes In Older Adults (NCT02448706), a randomized na study. The primary outcome is Speech Intelligibility Score.

2. Clinical Question

3. Trial Design

The registry identifies NCT02448706 as a completed, randomized, double-masked crossover study sponsored by Northwestern University. The record lists the study start as 2015-07 and primary completion as 2017-07. Its title is Variability In Hearing Aid Outcomes In Older Adults, and its official title is Characterizing Variability in Hearing Aid Outcomes Among Older Adults. The listed phase is NA.

A crossover design evaluates conditions in sequence within a study framework. As a general statistical matter, observations from the same participant can be related, so an analysis may need to account for that dependence. The registry identifies the model as crossover, but does not report the analysis used to account for participant pairing, treatment period, or sequence. It also does not report a washout interval or explain how any carryover between fitting periods was addressed. Those details should not be inferred from the design label alone.

The record names Hearing Aid Fitting A and Hearing Aid Fitting B as device interventions. The arm descriptions instead identify Hearing Aid Fitting Order A as high-level signal manipulation for 6 weeks followed by low-level manipulation for 6 weeks, and Hearing Aid Fitting Order B as low-level manipulation for 6 weeks followed by low-level manipulation for 6 weeks. The latter description does not distinguish the periods by signal level. The registry record does not clarify this wording, so the sequence labels cannot be used to reconstruct a more detailed protocol.

Randomization and double masking are recorded design features, but the methods used to generate or conceal the allocation and maintain masking are not reported in the registry record. The primary endpoint is speech intelligibility at 6 weeks. Secondary outcomes are Effectiveness of Aural Rehabilitation (EAR), the Speech and Spatial Qualities of Hearing (SSQ) Questionnaire, and adherence; each is listed with a 6-week timeframe. The posted outcome summaries use 40 observations for each signal-manipulation condition. The record does not explain whether these counts reflect complete paired observations or another analysis population.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Speech Intelligibility Score, measured over 6 weeks.

The primary endpoint is the Speech Intelligibility Score, reported as a mean and standard deviation. Participants listen to sentences presented against background noise and repeat what they hear. The score is the percentage of correctly repeated words, so it represents performance on a speech-under-noise task rather than a broad measure of hearing ability or everyday communication.

This type of endpoint is used to quantify how accurately speech is understood under a specified listening condition. Reporting a mean gives the average score for a condition, while the standard deviation describes how much scores vary around that average. Both are relevant in a study whose stated aim is to characterize variability in hearing-aid outcomes. A difference in group means describes a difference in average performance, but does not show whether every participant had a similar response or how individual responses changed between fitting conditions.

The registry gives a 6-week timeframe for the primary measure. Its posted means are 69.0 for high-level signal manipulation and 78.0 for low-level signal manipulation, with standard deviations of 18.0 and 15.0, respectively. The listed mean difference is 9.0. Those values describe the posted summary comparison. The record does not report the scoring range, a threshold for a practically important difference, or participant-level changes. Consequently, the size of the mean difference cannot by itself establish whether it would be important in daily life.

7. Results Posted in the Registry

• Low Level of Signal Manipulation: mean 78.00 versus 69.00 on control; difference 9.00 (95% CI 1.74 to 16.26), standardized effect d = 0.54; z-test p = 0.015.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Effectiveness of Aural Rehabilitation (EAR)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Hearing Aid Fitting B: mean 74.20 versus 60.00 on control; difference 14.20 (95% CI 4.48 to 23.92), standardized effect d = 0.64; z-test p = 0.004.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Hearing Aid Fitting B than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Speech and Spatial Qualities of Hearing (SSQ) Questionnaire

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Hearing Aid Fitting B: mean 6.95 versus 6.47 on control; difference 0.48 (95% CI -0.07 to 1.03), standardized effect d = 0.38; z-test p = 0.088.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Hearing Aid Fitting B and the control. The interval gives the range of effects that remain compatible with the data.

Adherence

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Hearing Aid Fitting B: mean 9.00 versus 9.00 on control; difference 0.00 (95% CI -1.56 to 1.56), standardized effect d = 0.00; z-test p = 1.000.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Hearing Aid Fitting B and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover different aspects of the hearing-aid experience: effectiveness of aural rehabilitation, perceived speech and spatial qualities of hearing, and adherence measured in hours per day. Across the posted summaries, the EAR and SSQ means are higher for Hearing Aid Fitting B than for Hearing Aid Fitting A, while the adherence means are equal. This directionally mixed pattern gives context to the primary speech intelligibility result, but the outcomes measure different constructs and are not interchangeable.

The registry reports an unadjusted comparison for EAR and SSQ, but does not list reported analyses for these outcomes. No comparison is listed for adherence beyond the posted means and its unadjusted summary statistics. These secondary measures can help describe whether the primary pattern is accompanied by differences in perceived benefit or use, but they do not establish a single consistent effect across all dimensions. Nor does agreement in direction between outcomes prove that they measure the same underlying change.

Unadjusted secondary results carry less weight as confirmatory evidence when the analysis plan and multiplicity strategy are not reported. In general, examining multiple outcomes creates more opportunities for an apparently notable result, even when no single outcome was selected in advance as a confirmatory test. The record does not say whether secondary endpoints were prespecified for formal hypothesis testing, whether any adjustment was planned, or how missing secondary measurements were handled.

The registry’s stated research goal concerns whether hearing-aid signal processing should be guided in part by individual cognitive characteristics. The posted secondary outcomes do not report a cognitive-characteristics analysis or a test of whether such characteristics modify response. They therefore offer descriptive context about rehabilitation, perceived hearing quality, and adherence, rather than direct evidence on that broader goal. The pattern should be treated as exploratory in the absence of a reported analysis plan and multiplicity approach.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Effectiveness of Aural Rehabilitation (EAR)
• Speech and Spatial Qualities of Hearing (SSQ) Questionnaire
• Adherence

9. Statistical Interpretation

10. Statistical Methods Explained

Statistical methods help separate the observed data from the assumptions needed to interpret them. The registry reports continuous outcomes as means with standard deviations. It also lists mean differences, confidence intervals, and unadjusted p-values for the primary comparison and two secondary comparisons. These summaries are useful for describing average outcomes and their uncertainty, but the record does not report a complete statistical analysis plan or sponsor-reported analyses.

What does a mean and standard deviation describe?

In general, a mean is the arithmetic average of measured values in a group. It is a compact description of the group’s central level, but it does not show the full distribution. A standard deviation describes the spread of individual measurements around the mean. A larger standard deviation generally indicates more variability in observed scores. The standard deviations for speech intelligibility are 18.0 for high-level manipulation and 15.0 for low-level manipulation. They describe between-observation spread in the posted summaries, not the variability of each participant’s change between conditions.

The registry’s primary score is the percentage of correctly repeated words in sentences presented against background noise. A mean difference of 9.0 expresses the contrast between the posted low-level and high-level means. It is an absolute difference on the endpoint scale, not a relative percentage change. The registry supplies no threshold for an important difference, so statistical estimates alone cannot determine practical importance.

What does a confidence interval add?

In general, a confidence interval describes the uncertainty around an estimated effect using a specified method and its assumptions. It is not a range that guarantees the true effect lies within it, nor is it a probability statement about the effect itself. For the primary comparison, the listed interval is 1.739 to 16.26. Its width reflects uncertainty in the estimate; the interval’s interpretation depends on the calculation being appropriate for the study design and data.

The interval is centered around the listed mean difference of 9.0, subject to rounding. Its endpoints are both above zero. This gives a different kind of information from the point estimate: the estimate identifies the observed contrast, while the interval indicates the range of values compatible with the calculation. The interval does not account for unreported sources of uncertainty, such as the consequences of an inappropriate independence assumption or unaddressed period effects.

What does an unadjusted p-value mean?

In general, a p-value describes how unusual data at least as discrepant as those observed would be under a specified null model and its assumptions. It is not the probability that a result happened by chance, and it does not quantify clinical importance. The primary unadjusted p-value is 0.015. The registry does not report the analysis plan that would establish whether this comparison was prespecified, which model generated the p-value, or how it handled the crossover structure. The value should therefore be interpreted as the posted unadjusted comparison rather than as a complete account of the inferential analysis.

An unadjusted result also does not account for testing multiple outcomes or comparisons. If multiple hypotheses are examined, the chance of at least one apparently unusual result can change. The record does not report a multiplicity adjustment or strategy. This is particularly relevant when interpreting secondary outcomes; their unadjusted p-values are not a substitute for a prespecified confirmatory plan.

Why does the crossover structure matter?

In general, crossover studies can compare conditions within participants, which can reduce the influence of stable differences between people. The trade-off is that measurements from the same person are related. Appropriate analysis may use the paired structure and may need to account for treatment period, sequence, or carryover. The registry labels the design crossover, but does not report whether the comparison uses paired observations or how period and sequence were handled. The listed counts of 40 for each condition do not establish that the same 40 people contributed to both summaries.

The record also contains sequence descriptions that are not fully clear: Order A is described as high-level manipulation followed by low-level manipulation, while Order B is described as low-level manipulation followed by low-level manipulation. No further clarification is reported. This limits the ability to assess how the stated sequence structure relates to the outcome comparison. The original design assumptions, including the planned analysis and any handling of missing outcomes, are not reported in the registry record.

These distinctions matter because a descriptive contrast between condition means is not necessarily equivalent to an analysis that models the crossover design. Without a reported analysis method, it is not possible to determine whether the point estimate, standard error, confidence interval, and p-value reflect within-participant dependence. The listed discrepancy indicator is false, but that does not supply the missing protocol details or establish which model was used.

11. Confidence Intervals and Precision

Precision concerns how narrowly the effect is estimated. For the primary speech intelligibility comparison, the mean difference is 9.0, with a confidence interval from 1.739 to 16.26. This interval expresses uncertainty around the estimate under the method’s assumptions. Its limits are separated by a noticeable range, so the estimate is not pin-point precise. The range remains on the positive side of zero, but that feature does not resolve questions about the crossover analysis or practical importance.

Group standard deviations also help describe the data’s variability: 18.0 for high-level manipulation and 15.0 for low-level manipulation. They do not provide the standard deviation of paired differences, which would be useful for understanding within-person variation in a crossover comparison. The registry does not report that quantity or participant-level data. It also does not report confidence intervals for the two condition means separately.

Confidence intervals are sensitive to both the amount of information and the variability in the data. In general, more precise estimates have narrower intervals, while greater uncertainty produces wider intervals. Here, the interval should be read alongside the design label and the absence of a reported crossover analysis method. Its numerical width alone cannot show whether the assumptions used to calculate it match the study’s data structure.

The posted interval is an uncertainty measure for the listed comparison, not a guarantee about future studies or individual outcomes. It also does not describe the range of likely participant responses. The standard deviations and the confidence interval answer different questions: the former describe spread of observed scores, while the latter describes uncertainty around an estimated mean difference. Neither establishes a threshold of practical significance, which is not reported in the registry record.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size figures are planning illustrations based on fractions of the observed effect, not the trial’s original design calculation. The registry record does not report the original sample-size assumptions or justification. For continuous outcomes, the figures are participants per arm, not a total number of participants. The record’s note that time-to-event results would use total events does not apply to these continuous outcomes.

The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect. For each such effect, the 80% power column means the estimated participants per arm for a future study to have an 80% chance of detecting an effect of that size under the planning assumptions, if the assumed effect and model are appropriate. The 90% power column gives the corresponding planning number for a 90% chance.

For the primary endpoint, planning for 100% of the observed effect gives 54 participants per arm for 80% power and 72 for 90% power. Planning for 75% of the observed effect gives 95 participants per arm for 80% power and 127 for 90% power. Planning for 50% of the observed effect gives 213 participants per arm for 80% power and 285 for 90% power. Smaller effects need more participants to detect under otherwise unchanged assumptions, which is why the planning figures rise as the target fraction falls.

These are illustrative calculations, not observed power and not evidence that the completed study had a particular chance of success. The registry does not report that these assumptions were used to plan the trial. The record also does not state whether the illustration accounts for crossover pairing, period effects, or other features of the recorded design. Planning figures should therefore be read as conditional examples, not as a reconstruction of the study’s original design.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = 9.000; pooled standard deviation σ = 16.568.
2. n = 2 × (1.96 + 0.842)² × 16.568² / 9.000² = 53.2, rounded up to 54 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- 9                      # observed difference in means
sigma <- 16.5680415257809                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(2448706)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 54 (80% power), 72 (90% power)
 75% of observed effect: n per arm = 95 (80% power), 127 (90% power)
 50% of observed effect: n per arm = 213 (80% power), 285 (90% power)
power.t.test: 54.2 
Simulated power at n = 54 per arm: 0.799 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several outcomes, comparisons, or analyses are examined. In general, the more tests performed, the greater the chance that at least one result will appear unusual under a collection of null hypotheses. Statistical plans can address this by defining a primary comparison, limiting confirmatory tests, or using an adjustment strategy. The registry identifies a primary outcome and three secondary outcomes, but does not report a multiplicity plan or adjustment. The unadjusted secondary p-values should therefore not be read as if they were protected by a family-wise testing strategy.

The record lists no sponsor-reported analyses. It does not give the prespecified statistical analysis plan, identify the exact model used for the posted comparisons, or state how the crossover pairing was handled. It also does not report whether period, sequence, or potential carryover effects were included. These omissions prevent a reader from determining whether the inferential calculations correspond to the randomized crossover design.

Missing-data methods and analysis populations are also not reported in the registry record. The study enrollment is 49, while the posted means use 40 observations per condition. The record does not explain why those figures differ, whether observations were missing, whether the outcome summaries include all randomized participants, or whether the condition summaries represent paired observations. These quantities should not be treated as interchangeable measures of the same analysis population.

The registry does not state whether an intention-to-treat, per-protocol, or other analysis population was used. It does not report how incomplete periods, withdrawals, or unavailable endpoint measurements were handled. These details matter particularly for a crossover study, where having a measurement under one condition but not the other can affect which participants contribute to a paired comparison. The posted summaries remain informative as descriptions, but the missing analysis and population details limit assessment of possible bias and reproducibility.

15. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The registry reports serious adverse event and other adverse event figures separately for Hearing Aid Fitting A and Hearing Aid Fitting B. For each fitting, 49 participants are listed as at risk for serious events and 0 as affected; 49 are listed as at risk for other events and 0 as affected. Deaths are reported as 0 for each fitting. No serious-event terms are listed.

These entries mean that the registry’s posted counts show zero affected participants in each of the reported event categories for both fitting labels. They do not establish that no symptoms or events of any kind occurred, and they do not describe the nature of monitoring, event collection, or attribution. The registry does not report further safety-analysis details or explain the relationship between these at-risk counts and the outcome summaries.

Zero recorded events in a study does not prove that the risk of an event is zero. In general, rare events may not be observed in a study of limited size, and an absence of recorded events is distinct from evidence that an intervention cannot cause harm. These figures should be read as the registry’s reported event counts, not as a comprehensive estimate of safety in other populations or settings.

16. What the Result Does, and Does Not, Mean

The posted primary comparison describes higher average speech intelligibility in the low-level signal-manipulation condition than in the high-level condition. The mean difference is 9.0, with a confidence interval from 1.739 to 16.26 and an unadjusted p-value of 0.015. These statistics characterize the observed summary comparison; they do not establish that every participant had a higher score in the low-level condition or that the difference would be the same in another sample.

The result is specific to a speech-in-noise task in which participants repeated sentences, with performance reported as the percentage of correctly repeated words. It should not be treated as a direct measure of every aspect of hearing, communication, satisfaction, or daily functioning. The secondary outcomes broaden the description, but they are not a substitute for the primary endpoint and do not provide a direct report of cognitive characteristics as predictors of response.

The study is described as randomized, double-masked, and crossover, but the registry does not report the full statistical method, treatment-period analysis, or handling of sequence and carryover. The unusual sequence wording for Order B is not clarified. Because the posted comparison’s relationship to the crossover design is not reported, the confidence interval and p-value should be interpreted with that limitation in mind.

The result does not establish a clinically important threshold, a universal preferred fitting, or a basis for individual treatment selection. The registry record does not report a threshold for meaningful change, a cognitive subgroup analysis, or an analysis connecting cognitive characteristics with outcome variation. Its stated goal is to investigate whether processing should be guided partly by cognitive characteristics; the posted results do not answer that broader question.

In general, statistical evidence and practical importance are separate. An estimate can be statistically distinguishable from zero without establishing that its size matters to patients, and uncertainty around an estimate remains relevant even when a p-value is small. The registry’s posted data are best used to understand the reported averages, variability, and limits of the available analysis rather than to make clinical decisions.

17. Why This Trial Matters Statistically

This study is statistically interesting because it combines a research question about variation in hearing-aid outcomes with a crossover design. Comparing conditions within a participant can, in general, reduce the influence of stable differences between participants. But that advantage depends on appropriate analysis of paired measurements and careful handling of period and sequence. The registry identifies a crossover design without reporting those analytic details, making the distinction between design and analysis especially important.

The primary endpoint is a continuous measure with group means and standard deviations, allowing readers to consider both average performance and observed spread. The posted standard deviations show that scores vary within each condition. However, group-level spreads do not reveal how much each person’s score changes between conditions, which is central to a question about individual variability. Participant-level changes and a reported model of that variation are not available in the registry record.

The registry’s stated goal is to explore whether cognitive characteristics should help guide signal processing. That question concerns possible differences in response across people, not only an average contrast between conditions. The posted primary comparison addresses average speech intelligibility, while the listed outcomes do not report a cognitive-characteristics analysis. It is therefore important not to confuse an overall mean difference with evidence about who benefits from which processing approach.

Finally, this record illustrates how uncertainty about analysis populations and missing methods affects interpretation. Enrollment is 49, whereas each condition’s posted outcome summary lists 40 observations. Without an explanation of the relationship, paired completeness, and analysis set, readers cannot reconstruct the comparison. The statistical contribution is consequently as much a reminder to examine design-to-analysis alignment as it is a report of a mean contrast.

18. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

Several limitations are specific to the information reported for this trial. The most important is that the registry labels the study as a crossover but does not describe the analysis model. It is not reported whether the comparison accounts for measurements from the same participants, treatment period, assigned sequence, or carryover. These are not minor details in a design where conditions are given in sequence; they determine how estimates and uncertainty should be calculated.

The two sequence descriptions are not fully informative. Order A is described as high-level signal manipulation followed by low-level signal manipulation, while Order B is described as low-level manipulation followed by low-level manipulation. The registry record does not clarify this wording. It is therefore not possible to reconstruct the actual contrast and sequence structure beyond what is recorded.

The enrollment is 49, but each posted condition summary lists 40 observations. The record does not explain the difference, identify the analysis population, or state whether both condition measurements were available for the same participants. Missing-data handling and reasons for incomplete data are not reported. This limits evaluation of selection effects and of whether the reported comparison represents all randomized participants.

The record does not report a protocol-level statistical plan, sample-size justification, multiplicity strategy, or prespecified analysis details. It lists no reported analyses. The listed unadjusted comparisons therefore cannot be checked against a documented model or analysis hierarchy. The sample-size illustrations are based on observed effects and do not fill that gap.

Finally, the primary endpoint is a speech-in-noise task, and the registry does not report a threshold for meaningful change or a cognitive-characteristics analysis. The posted summaries cannot establish individual-level variation in response or the project’s broader proposed link between cognition and processing. These limitations concern what can be concluded from the registry record; they do not establish that unreported procedures were not carried out.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Variability In Hearing Aid Outcomes In Older Adults — NCT02448706.
• Registry record last updated 2020-11-17; status last verified 2020-10.
• Results first posted 2020-01-14.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

22. Record Summary

NCT02448706 is a completed randomized, double-masked crossover study sponsored by Northwestern University, with a recorded enrollment of 49. The primary outcome is mean speech intelligibility at 6 weeks, measured as the percentage of correctly repeated words in sentences presented against background noise.

The posted primary means are 69.0 for high-level signal manipulation and 78.0 for low-level signal manipulation. The listed mean difference is 9.0, with a confidence interval from 1.739 to 16.26 and an unadjusted p-value of 0.015. Three secondary outcomes are also posted: EAR, SSQ, and adherence.

The registry record does not report the full statistical analysis plan, how crossover pairing and sequence were handled, the analysis population, or missing-data methods. It also does not clarify the sequence description for Order B. The posted comparisons should therefore be read as summary results with important analysis details unreported.