This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Evaluation of a HiRes™ Optima Sound Processing Strategy for the HiResolution™ Bionic Ear (NCT01616576), a randomized na study. The primary outcome is Speech Perception With Control and Experimental Conditions Both Tested in Quiet, in Speech-spectrum Noise, and in Multi-talker Babble Noise..
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT01616576 |
| Phase | NA |
| Condition | Severe to Profound Hearing Loss, in Adult Users of Advanced Bionics HiResolution™ Bionic Ear System |
| Allocation | Randomized |
| Design model | Crossover |
| Masking | None (open label) |
| Enrollment | 36 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Advanced Bionics |
| Results posted | Yes |
2. Clinical Question
Population
Adults with severe to profound hearing loss who use the Advanced Bionics HiResolution™ Bionic Ear System. The registry lists enrollment of 36 participants. It does not provide further eligibility details or a demographic profile in the record.
Intervention
The experimental condition was the HiRes™ Optima sound processing strategy for the HiResolution™ Bionic Ear System. Participants assigned to the experimental-first sequence used it during the first week; participants in the other sequence used it during the second week.
Comparator
The comparator was the current, or Control, sound processing strategy. In the Control-first sequence it was used in the first week, followed by the experimental strategy. In the other sequence it followed the experimental strategy.
Primary question
The study aimed to compare sentence recognition scores with the two strategies, using the AzBio corpus. Testing was described in quiet, speech-spectrum noise, and multi-talker babble noise. The planned score contrast was Experimental minus Control, with data from both sequence groups pooled for analysis.
The central statistical question is whether measured AzBio percent-correct scores differ between sound processing strategies when the same participants experience both conditions. This is a within-participant comparison: each participant’s score under one strategy can be considered alongside that participant’s score under the other. That structure differs from a simple comparison of unrelated groups, because a participant’s general level of sentence recognition can influence both measurements.
The study randomized the order in which participants received the conditions, rather than assigning participants to receive only one strategy. The registry describes one week of use per condition and a total primary-outcome timeframe of 2 weeks. It does not report a numerical estimate of the score contrast, its uncertainty, or a test of the primary question. Therefore, the record describes the research question and measurement plan, but does not establish the size or precision of any observed difference.
3. Trial Design
This was a randomized, unmasked, crossover device study. The registry lists enrollment of 36 and describes two treatment sequences. Group A used Control first and Experimental second; Group B used Experimental first and Control second. Each condition was used for one week, giving a 2-week primary-outcome timeframe. The recorded study status is completed, with study start in 2012-05 and primary completion in 2012-09.
In a crossover design, participants act as their own comparison by receiving both conditions at different times. Randomizing the order helps distribute order-related influences across sequences. It does not, by itself, rule out period effects, changes over time, or carryover from the first condition into the second. The registry record does not report whether a washout period was used, how carryover was assessed, or whether period and sequence effects were included in the statistical model.
The study was unmasked, meaning the registry does not describe masking of participants or other study personnel. For a sound processing strategy, participants may be aware of which condition they are using, but the extent to which awareness affected testing or scoring is not reported in the registry record. The record also does not specify who administered or scored the AzBio tests.
The primary outcome description says data from Group A and Group B were pooled. Pooling sequence groups is compatible with an overall within-participant comparison, but the registry does not give the analysis method or explain how the paired measurements and treatment order were handled. It also does not provide participant flow, exclusions, adherence, or the number with evaluable measurements in each testing condition. As a result, the design can be characterized from the assigned sequences and use periods, while important implementation and analysis details remain unavailable.
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
| Stage | Control First, Then Experimental (Group A) | Experimental First, Then Control (Group B) |
|---|---|---|
| STARTED | 18 | 18 |
| COMPLETED | 18 | 18 |
| NOT COMPLETED | 0 | 0 |
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.
| Characteristic | Unit | Control First, Then Experimental (Group A) | Experimental First, Then Control (Group B) | Total |
|---|---|---|---|---|
| Participants analysed | 18 | 18 | 36 | |
| Age, Categorical: <=18 years | Participants | 0 | 0 | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 10 | 11 | 21 |
| Age, Categorical: >=65 years | Participants | 8 | 7 | 15 |
| Age, Continuous | years | 62.0 (SD 14.81) | 58.9 (SD 13.86) | 60.4 (SD 14.22) |
| Sex: Female, Male: Female | Participants | 10 | 9 | 19 |
| Sex: Female, Male: Male | Participants | 8 | 9 | 17 |
| Region of Enrollment: United States | participants | 18 | 18 | 36 |
6. Primary Endpoint
The registry lists the primary outcome as Speech Perception With Control and Experimental Conditions Both Tested in Quiet, in Speech-spectrum Noise, and in Multi-talker Babble Noise., measured over 2 weeks.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | Standard Deviation |
| Unit | Percent Correct |
| Endpoint classification | Continuous Endpoint |
The primary endpoint was sentence recognition measured as percent correct with the AzBio corpus. The registry describes testing with the Control and Experimental sound processing strategies in quiet, speech-spectrum noise, and multi-talker babble noise. The corpus is described as containing 33 lists of 20 sentences each, with 6 to 10 words per sentence, and lists equated for intelligibility. The endpoint timeframe was 2 weeks.
Percent correct represents the proportion of tested sentence material recognized correctly, expressed on a percentage scale. Comparing scores under the two strategies is intended to quantify how sentence recognition differed across the conditions in the tested listening environments. The registry specifies the direction of the contrast as Experimental AzBio scores minus Control AzBio scores. A positive contrast would therefore indicate a higher score under Experimental in the stated subtraction, while a negative contrast would indicate a lower score; that interpretation describes the sign convention, not an observed result.
Speech-in-noise testing can capture performance under listening conditions that differ from quiet. The endpoint is clinically relevant to the study’s stated comparison because it directly measures sentence recognition rather than an indirect device characteristic. However, the registry does not report numerical scores, variability, or separate estimates by listening condition. It also does not identify a single summarized comparison across the listed conditions. The outcome description therefore explains what was to be measured and compared, but not the result.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record supports a description of the trial’s question and design, but not an estimate of the primary outcome. It identifies AzBio sentence recognition, percent correct, the two sound processing strategies, the testing environments, and the intended direction of subtraction. It does not report the primary outcome values, a mean difference, a confidence interval, a test statistic, or a p-value. No conclusion about the magnitude or precision of the score contrast can be drawn from the record’s outcome information.
The distinction between a planned comparison and a reported result is important. The endpoint description says Experimental scores minus Control scores would be used and that both sequence groups were pooled. That states the intended contrast; it does not indicate whether the contrast was positive, negative, close to zero, or statistically distinguishable from zero. The primary outcome has several categories or time points and is not reduced to one comparison in the available record. Thus, even the overall result cannot be represented by one reported estimate based on these registry entries.
The crossover design offers a natural paired comparison because each participant used both strategies. In general, paired analyses account for the relationship between measurements from the same person and can be more informative than treating them as independent. But the registry does not say whether a paired analysis was used, how missing measurements were handled, whether period effects were considered, or whether the result was adjusted for any other factors. The statistical analysis plan is not reported in the registry record, and no analyses are listed.
There is also no reported uncertainty measure to show how precisely any estimated contrast was determined. A confidence interval, in general, describes a range of values compatible with the data and the method used; without one, the extent of uncertainty cannot be assessed. The record’s lack of a numerical estimate is not evidence that the strategies had identical scores, nor does it establish a difference. It means the result is unavailable in the registry record.
Interpretation should also account for the three stated listening environments. Scores from quiet, speech-spectrum noise, and multi-talker babble may represent distinct outcome categories. The registry does not provide category-specific results or identify a single combined score. A result for one listening environment could not be assumed to represent the others, and the record does not permit a comparison across environments.
Finally, the figures for adverse events are separate from the efficacy question. They describe affected participants by sequence group, but do not supply information about speech-recognition performance. The safety record lists no serious adverse events among the participants at risk in either group and lists other adverse events for both groups. Those observations should be reported as recorded, without using them to infer the primary outcome or a causal safety comparison.
9. Statistical Methods Explained
The registry describes a randomized crossover comparison, but it does not report the statistical analysis methods used. The distinction matters: the study design tells how participants were assigned to the order of conditions, while the analysis plan specifies how measurements are converted into estimates and how uncertainty is quantified. The record lists no analyses, so the methods below explain general principles relevant to this design rather than claiming that any particular procedure was used.
What is the treatment contrast?
The planned contrast is Experimental AzBio percent-correct score minus Control AzBio percent-correct score. For each participant with measurements under both conditions, a difference can be defined by subtracting that participant’s Control score from the Experimental score. A paired summary then describes the average of those within-participant differences. This approach preserves the crossover structure: the comparison is based on changes within people rather than on separate sets of people receiving only one condition.
The sign of a difference depends on the subtraction order. Under the registry’s stated order, positive values mean the Experimental score is higher than the Control score, and negative values mean it is lower. This is a general interpretation of the specified contrast, not a claim about the result. No estimated contrast is reported in the registry record.
Why does the crossover structure matter?
Measurements from the same participant are usually related. A person who tends to recognize more sentences may score relatively highly under both strategies. A paired method accounts for this relationship by evaluating the within-person contrast. Treating the two condition measurements as if they came from unrelated people would ignore the pairing and could give an inappropriate estimate of uncertainty.
Randomizing the order of conditions helps prevent one sequence from being deliberately assigned to a particular kind of participant. However, a crossover comparison can also be influenced by period effects, such as changes between the first and second week, and carryover, where experience with the first condition affects measurement during the next condition. These are general design considerations. Whether this trial used a washout period or modeled period, sequence, or carryover effects is not reported in the registry record.
What do the descriptive statistics mean?
The endpoint is continuous in the sense that percent-correct scores can take a range of values. A mean describes the average score or average paired difference. A standard deviation describes how spread out participant measurements are around their mean. These quantities answer different questions: the mean describes the center of the observed scores, while the standard deviation describes participant-to-participant variation. The registry identifies mean and standard deviation as the measure and dispersion, but supplies no outcome values.
For paired data, the spread of individual differences is especially relevant to uncertainty in the average contrast. Separate standard deviations for the Experimental and Control conditions alone do not fully describe the variability of paired differences; the within-participant relationship also matters. The registry does not report the needed values or specify how dispersion was summarized in an analysis.
How should uncertainty be assessed?
In general, a confidence interval around a mean difference conveys the precision of that estimate. A narrower interval indicates greater precision under the assumptions of the method, while a wider interval indicates more uncertainty. A significance test addresses a related but different question: whether the data are inconsistent with a specified null hypothesis under a chosen model. Neither a confidence interval nor a test result establishes that an observed difference is important in practice. No interval, significance test, or threshold for an important difference is reported in the registry record.
For a crossover trial, the analysis should reflect the paired nature of the data and the sequence structure. The exact method could depend on the data distribution, completeness, and design assumptions. The registry does not state whether the analysis used a paired test, a repeated-measures model, a mixed model, or another procedure. It also does not report assumptions or diagnostic checks. None should be inferred from the endpoint description alone.
Why can multiple conditions complicate interpretation?
The endpoint description names quiet, speech-spectrum noise, and multi-talker babble noise. If each condition produces a separate contrast, there may be several related estimates rather than one. In general, conducting multiple tests can increase the chance of at least one apparently notable finding when all comparisons are considered together. A prespecified primary comparison or a multiplicity adjustment can help define how such results should be interpreted. Whether a particular strategy was prespecified or applied is not reported in the registry record.
The record marks the primary outcome as having several categories or time points and does not reduce it to one comparison. It does not provide the category-specific estimates, specify a hierarchy among conditions, or state whether any adjustment for multiple comparisons was planned. Consequently, neither the number of inferential comparisons nor their statistical interpretation can be reconstructed from the registry description.
What do planning illustrations represent?
Sample-size planning asks how many participants may be needed under stated assumptions to detect a chosen effect with a specified operating characteristic. A planning illustration based on an observed effect is not the same as the original trial’s design justification and is not observed power. The stated illustration assumptions use a two-sided significance level of 0.05, equal allocation between arms, a normal approximation, and an effect defined as a fraction of the observed effect. For non-time-to-event outcomes, the planning quantities are participants per arm. These are planning assumptions, not trial findings.
The observed primary effect and the trial’s own sample-size justification are not reported in the registry record. Therefore, a numerical planning interpretation cannot be independently grounded in the outcome details. No claim about whether the trial was adequately sized, or what effect it could reliably detect, follows from enrollment alone.
10. Multiplicity, Missing Data and Analysis Populations
The registry describes the primary outcome across quiet, speech-spectrum noise, and multi-talker babble noise, and identifies a 2-week timeframe. It also indicates that the primary outcome has several categories or time points and is not reduced to one comparison. The record does not provide separate results by category, define one category as the sole primary comparison, or describe a multiplicity strategy. Whether testing across listening environments was handled with a prespecified hierarchy, adjusted significance threshold, or another approach is not reported in the registry record.
Multiplicity matters when several related comparisons are assessed. As a general principle, the more inferential comparisons considered, the greater the opportunity for a result to appear noteworthy by chance if each is judged separately. Adjustment methods and outcome hierarchies address that issue in different ways. Without the analysis plan and category-specific results, readers cannot determine how multiplicity was considered or how any possible findings should be interpreted together.
Missing-data procedures are also not reported. The registry does not state how many participants completed each testing condition, whether any measurements were unavailable, why data might be missing, or whether an imputation or other method was used. In a crossover study, a participant may have one condition’s measurement but not the other; that affects whether a paired contrast can be calculated. The record does not explain how such cases would enter an analysis.
The analysis population is not defined. Although the record states enrollment of 36 and says both sequence groups were pooled for the primary analysis, it does not specify whether all randomized participants, only participants completing both periods, or another group was included. The number of participants included in the primary outcome analysis is not reported in the registry record. These omissions limit assessment of attrition, selection into the analysis, and the relationship between the randomized design and the reported estimate.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Control First, Then Experimental (Group A) | 0 | 18 | 0.0% |
| Experimental First, Then Control (Group B) | 0 | 18 | 0.0% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Control First, Then Experimental (Group A) | 2 of 18 |
| Experimental First, Then Control (Group B) | 3 of 18 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Control First, Then Experimental (Group A) | Experimental First, Then Control (Group B) |
|---|---|---|---|
| Headache | General disorders | 1 of 18 | 3 of 18 |
| Cold | General disorders | 1 of 18 | 0 of 18 |
The registry’s serious adverse-event figures list 0 affected participants among 18 at risk in Group A and 0 among 18 at risk in Group B. The record lists other adverse events for 2 participants in Group A and 3 in Group B, with 18 at risk in each group. These are counts of affected participants as recorded, not a complete description of event severity, duration, cause, or relationship to the device.
Because this is a crossover study, the groups refer to treatment sequences: Control first, then Experimental, and Experimental first, then Control. The adverse-event figures are reported by sequence group, not separately by the condition being used when an event occurred. The registry record therefore does not support attributing the other adverse events to one sound processing strategy or making a comparative risk estimate.
The deaths field has no numerical value reported in the registry record. The adverse-event figures should be read as descriptive counts, not as proof that an event could not occur or as a formal test of safety differences. They also do not answer the primary speech-recognition question.
12. What the Result Does, and Does Not, Mean
The record establishes that a randomized, unmasked crossover study was completed to compare Control and HiRes™ Optima sound processing strategies in adult users of the HiResolution™ Bionic Ear System. It describes one week of use for each condition, two treatment sequences, and AzBio sentence recognition testing across quiet and noise environments. It also states that the intended contrast was Experimental minus Control and that data from both sequences were pooled.
Those design details clarify what the investigators set out to compare. They do not supply the comparison’s numerical result. The registry record does not report the mean scores, the mean difference, its standard deviation, a confidence interval, or a statistical test. There is therefore no basis in the record for describing the direction, size, or precision of the primary outcome. Absence of a reported estimate is not evidence of no difference, and it is not evidence that one strategy produced higher scores.
The stated endpoint covers multiple listening environments. Without results separated by condition or a stated combined outcome, it is not possible to know whether patterns differed between quiet, speech-spectrum noise, and multi-talker babble. Nor can one infer how participants responded individually from an average that is not reported. The statistical analysis details and analysis population are also unavailable, which prevents evaluation of how the crossover measurements were handled.
The record can still be useful for understanding the trial’s design and intended measurement. It makes clear that the comparison was within participants and that treatment order was randomized. But assessment of the observed effect requires outcome estimates and uncertainty measures, while assessment of robustness requires details about missing data, period effects, and analysis methods. Those details are not reported in the registry record.
The adverse-event counts provide a limited descriptive account by sequence group. They should not be interpreted as evidence about the primary endpoint or as a strategy-specific safety comparison, because the registry does not attribute events to the period or condition in which they occurred. Overall, the available information defines the question and design, but does not establish an efficacy or comparative safety conclusion.
13. Why This Trial Matters Statistically
This trial is statistically instructive because it uses a crossover design to compare two sound processing strategies. When participants receive both conditions, their own measurements can serve as a reference for one another. This can help separate differences between conditions from stable differences among participants, provided the analysis respects the paired structure and the design’s assumptions.
Randomizing sequence is important in a crossover study, but it does not resolve every source of bias. Changes between periods and carryover from one condition to the next can affect the contrast. The registry describes the order and duration of use, yet does not report whether these effects were assessed or how they were incorporated into the analysis. This illustrates why a design description alone is not a substitute for a statistical analysis plan.
The outcome also has multiple listening environments. Reporting distinct estimates can show whether the comparison is consistent across conditions, while a clearly defined primary contrast can make the main inference easier to interpret. In this record, results are not reported by environment and the outcome is not reduced to one comparison. The missing statistical details therefore matter not only for precision but also for identifying what the primary result would represent.
Finally, a completed study record can describe enrollment, intervention sequence, endpoints, and adverse-event counts without supplying the numerical outcome analysis. Readers need effect estimates and uncertainty, along with analysis-population and missing-data information, to assess what the data show. The record’s design information is informative, but it cannot replace those results.
14. 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.
The most consequential limitation of the registry record is that it does not report numerical primary-outcome results. There is no mean score under either condition, no estimated Experimental-minus-Control difference, and no measure of uncertainty. The record therefore cannot support a quantitative interpretation of sentence-recognition performance.
The endpoint is described across quiet, speech-spectrum noise, and multi-talker babble, but the available information does not give separate estimates for these environments or a single combined estimate. The primary outcome is not reduced to one comparison. This leaves the target of any overall inference unclear from the record alone.
Several design and analysis details are absent. The registry does not report whether a washout period was used, whether period or carryover effects were examined, or which statistical model was applied. It does not identify the analysis population, provide participant flow through both periods, or describe missing-data handling. The unmasked design may also be relevant to assessment and reporting, but the record does not explain who knew the condition during testing or scoring.
The adverse-event information is limited to counts by sequence group. It does not identify the condition in use when an event occurred, provide event descriptions, or report a numerical death count. The registry also does not report the trial’s sample-size justification or statistical analysis plan. These limitations restrict conclusions about precision, robustness, multiplicity, and comparative safety, even though the randomized crossover structure and intended endpoint are described.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Evaluation of a HiRes™ Optima Sound Processing Strategy for the HiResolution™ Bionic Ear — NCT01616576.
- Registry record last updated 2020-07-22; status last verified 2014-08.
- Results first posted 2014-08-15.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
NCT01616576 is a completed randomized, unmasked crossover device study sponsored by Advanced Bionics. The registry lists enrollment of 36 and describes two sequences, each using Control and HiRes™ Optima sound processing strategies for one week. The primary outcome was AzBio sentence recognition, measured as percent correct over a 2-week timeframe and tested in quiet, speech-spectrum noise, and multi-talker babble.
The intended contrast was Experimental scores minus Control scores, with the two sequence groups pooled. The registry record does not report numerical primary-outcome results, a statistical analysis, or an analysis population. It also does not describe missing-data procedures or how the multiple testing environments were handled. Serious adverse events were listed for 0 participants in each sequence group; other adverse events were listed for 2 in Group A and 3 in Group B, among 18 at risk in each. These safety counts are descriptive and are not assigned to a specific strategy.